Download ATmega32 - ickala.com

8-bit Microcontroller with 32KBytes In-System Programmable Flash

ATmega32A
Features• High-performance, Low-power Atmel®AVR® 8-bit Microcontroller• Advanced RISC Architecture
– 131 Powerful Instructions – Most Single-clock Cycle Execution– 32 × 8 General Purpose Working Registers– Fully Static Operation– Up to 16MIPS Throughput at 16MHz– On-chip 2-cycle Multiplier

• High Endurance Non-volatile Memory segments– 32Kbytes of In-System Self-programmable Flash program memory– 1024Bytes EEPROM– 2Kbytes Internal SRAM– Write/Erase Cycles: 10,000 Flash/100,000 EEPROM– Data retention: 20 years at 85C/100 years at 25C(1)

– Optional Boot Code Section with Independent Lock Bits• In-System Programming by On-chip Boot Program• True Read-While-Write Operation

– Programming Lock for Software Security• JTAG (IEEE std. 1149.1 Compliant) Interface

– Boundary-scan Capabilities According to the JTAG Standard– Extensive On-chip Debug Support– Programming of Flash, EEPROM, Fuses, and Lock Bits through the JTAG Interface

• Atmel QTouch® library support– Capacitive touch buttons, sliders and wheels– Atmel QTouch and QMatrix acquisition– Up to 64 sense channels

• Peripheral Features– Two 8-bit Timer/Counters with Separate Prescalers and Compare Modes– One 16-bit Timer/Counter with Separate Prescaler, Compare Mode, and Capture Mode– Real Time Counter with Separate Oscillator– Four PWM Channels– 8-channel, 10-bit ADC

• 8 Single-ended Channels• 7 Differential Channels in TQFP Package Only• 2 Differential Channels with Programmable Gain at 1x, 10x, or 200x

– Byte-oriented Two-wire Serial Interface– Programmable Serial USART– Master/Slave SPI Serial Interface– Programmable Watchdog Timer with Separate On-chip Oscillator– On-chip Analog Comparator

• Special Microcontroller Features– Power-on Reset and Programmable Brown-out Detection– Internal Calibrated RC Oscillator– External and Internal Interrupt Sources– Six Sleep Modes: Idle, ADC Noise Reduction, Power-save, Power-down, Standby and

Extended Standby• I/O and Packages

– 32 Programmable I/O Lines– 40-pin PDIP, 44-lead TQFP, and 44-pad QFN/MLF

• Operating Voltages– 2.7V - 5.5V

• Speed Grades– 0 - 16MHz

• Power Consumption at 1MHz, 3V, 25C– Active: 0.6mA– Idle Mode: 0.2mA– Power-down Mode: < 1µA

Atmel-8155D-AVR-ATmega32A-Datasheet_02/2014

1. Pin Configurations

Figure 1-1. Pinout ATmega32A

(XCK/T0) PB0(T1) PB1

(INT2/AIN0) PB2(OC0/AIN1) PB3

(SS) PB4(MOSI) PB5(MISO) PB6(SCK) PB7

RESETVCCGND

XTAL2XTAL1

(RXD) PD0(TXD) PD1(INT0) PD2(INT1) PD3

(OC1B) PD4(OC1A) PD5(ICP1) PD6

PA0 (ADC0)PA1 (ADC1)PA2 (ADC2)PA3 (ADC3)PA4 (ADC4)PA5 (ADC5)PA6 (ADC6)PA7 (ADC7)AREFGNDAVCCPC7 (TOSC2)PC6 (TOSC1)PC5 (TDI)PC4 (TDO)PC3 (TMS)PC2 (TCK)PC1 (SDA)PC0 (SCL)PD7 (OC2)

PA4 (ADC4)PA5 (ADC5)PA6 (ADC6)PA7 (ADC7)AREFGNDAVCCPC7 (TOSC2)PC6 (TOSC1)PC5 (TDI)PC4 (TDO)

(MOSI) PB5(MISO) PB6(SCK) PB7

RESETVCCGND

XTAL2XTAL1

(RXD) PD0(TXD) PD1(INT0) PD2

(IN

T1)

PD

3(O

C1B

) P

D4

(OC

1A)

PD

5(I

CP

1) P

D6

(OC

2) P

D7

VC

CG

ND

(SC

L) P

C0

(SD

A)

PC

1(T

CK

) P

C2

(TM

S)

PC

3

PB

4 (

SS

)P

B3

(A

IN1/

OC

0)P

B2

(A

IN0/

INT

2)P

B1

(T

1)P

B0

(X

CK

/T0)

GN

DV

CC

PA0

(A

DC

0)PA

1 (

AD

C1)

PA2

(A

DC

2)PA

3 (

AD

C3)

PDIP

TQFP/MLF

Note: Bottom pad shouldbe soldered to ground.

2ATmega32A [DATASHEET]Atmel-8155D-AVR-ATmega32A-Datasheet_02/2014

2. OverviewThe Atmel®AVR® ATmega32A is a low-power CMOS 8-bit microcontroller based on the AVR enhanced RISCarchitecture. By executing powerful instructions in a single clock cycle, the ATmega32A achieves throughputsapproaching 1 MIPS per MHz allowing the system designer to optimize power consumption versus processingspeed.

2.1 Block Diagram

Figure 2-1. Block Diagram

The Atmel®AVR® core combines a rich instruction set with 32 general purpose working registers. All the 32 regis-ters are directly connected to the Arithmetic Logic Unit (ALU), allowing two independent registers to be accessed in

INTERNALOSCILLATOR

OSCILLATOR

WATCHDOGTIMER

MCU CTRL.& TIMING

OSCILLATOR

TIMERS/COUNTERS

INTERRUPTUNIT

STACKPOINTER

EEPROM

SRAM

STATUSREGISTER

USART

PROGRAMCOUNTER

PROGRAMFLASH

INSTRUCTIONREGISTER

INSTRUCTIONDECODER

PROGRAMMINGLOGIC SPI

ADCINTERFACE

COMP.INTERFACE

PORTA DRIVERS/BUFFERS

PORTA DIGITAL INTERFACE

GENERALPURPOSE

REGISTERS

X

Y

Z

ALU

+-

PORTC DRIVERS/BUFFERS

PORTC DIGITAL INTERFACE

PORTB DIGITAL INTERFACE

PORTB DRIVERS/BUFFERS

PORTD DIGITAL INTERFACE

PORTD DRIVERS/BUFFERS

XTAL1

XTAL2

RESET

CONTROLLINES

VCC

GND

MUX &ADC

AREF

PA0 - PA7 PC0 - PC7

PD0 - PD7PB0 - PB7

AVR CPU

TWI

AVCC

INTERNALCALIBRATEDOSCILLATOR


one single instruction executed in one clock cycle. The resulting architecture is more code efficient while achievingthroughputs up to ten times faster than conventional CISC microcontrollers.

The ATmega32A provides the following features: 32Kbytes of In-System Programmable Flash Program memorywith Read-While-Write capabilities, 1024bytes EEPROM, 2Kbyte SRAM, 32 general purpose I/O lines, 32 generalpurpose working registers, a JTAG interface for Boundary-scan, On-chip Debugging support and programming,three flexible Timer/Counters with compare modes, Internal and External Interrupts, a serial programmableUSART, a byte oriented Two-wire Serial Interface, an 8-channel, 10-bit ADC with optional differential input stagewith programmable gain (TQFP package only), a programmable Watchdog Timer with Internal Oscillator, an SPIserial port, and six software selectable power saving modes. The Idle mode stops the CPU while allowing theUSART, Two-wire interface, A/D Converter, SRAM, Timer/Counters, SPI port, and interrupt system to continuefunctioning. The Power-down mode saves the register contents but freezes the Oscillator, disabling all other chipfunctions until the next External Interrupt or Hardware Reset. In Power-save mode, the Asynchronous Timer con-tinues to run, allowing the user to maintain a timer base while the rest of the device is sleeping. The ADC NoiseReduction mode stops the CPU and all I/O modules except Asynchronous Timer and ADC, to minimize switchingnoise during ADC conversions. In Standby mode, the crystal/resonator Oscillator is running while the rest of thedevice is sleeping. This allows very fast start-up combined with low-power consumption. In Extended Standbymode, both the main Oscillator and the Asynchronous Timer continue to run.

The device is manufactured using Atmel’s high density nonvolatile memory technology. The On-chip ISP Flashallows the program memory to be reprogrammed in-system through an SPI serial interface, by a conventional non-volatile memory programmer, or by an On-chip Boot program running on the AVR core. The boot program can useany interface to download the application program in the Application Flash memory. Software in the Boot Flashsection will continue to run while the Application Flash section is updated, providing true Read-While-Write opera-tion. By combining an 8-bit RISC CPU with In-System Self-Programmable Flash on a monolithic chip, the AtmelATmega32A is a powerful microcontroller that provides a highly-flexible and cost-effective solution to many embed-ded control applications.

The Atmel AVR ATmega32A is supported with a full suite of program and system development tools including: Ccompilers, macro assemblers, program debugger/simulators, in-circuit emulators, and evaluation kits.

2.2 Pin Descriptions

2.2.1 VCCDigital supply voltage.

2.2.2 GNDGround.

2.2.3 Port A (PA7:PA0)Port A serves as the analog inputs to the A/D Converter.

Port A also serves as an 8-bit bi-directional I/O port, if the A/D Converter is not used. Port pins can provide internalpull-up resistors (selected for each bit). The Port A output buffers have symmetrical drive characteristics with bothhigh sink and source capability. When pins PA0 to PA7 are used as inputs and are externally pulled low, they willsource current if the internal pull-up resistors are activated. The Port A pins are tri-stated when a reset conditionbecomes active, even if the clock is not running.

2.2.4 Port B (PB7:PB0)Port B is an 8-bit bi-directional I/O port with internal pull-up resistors (selected for each bit). The Port B output buf-fers have symmetrical drive characteristics with both high sink and source capability. As inputs, Port B pins that areexternally pulled low will source current if the pull-up resistors are activated. The Port B pins are tri-stated when areset condition becomes active, even if the clock is not running.


Port B also serves the functions of various special features of the ATmega32A as listed on page 56.

2.2.5 Port C (PC7:PC0)Port C is an 8-bit bi-directional I/O port with internal pull-up resistors (selected for each bit). The Port C output buf-fers have symmetrical drive characteristics with both high sink and source capability. As inputs, Port C pins that areexternally pulled low will source current if the pull-up resistors are activated. The Port C pins are tri-stated when areset condition becomes active, even if the clock is not running. If the JTAG interface is enabled, the pull-up resis-tors on pins PC5(TDI), PC3(TMS) and PC2(TCK) will be activated even if a reset occurs.

The TD0 pin is tri-stated unless TAP states that shift out data are entered.

Port C also serves the functions of the JTAG interface and other special features of the ATmega32A as listed onpage 58.

2.2.6 Port D (PD7:PD0)Port D is an 8-bit bi-directional I/O port with internal pull-up resistors (selected for each bit). The Port D output buf-fers have symmetrical drive characteristics with both high sink and source capability. As inputs, Port D pins that areexternally pulled low will source current if the pull-up resistors are activated. The Port D pins are tri-stated when areset condition becomes active, even if the clock is not running.

Port D also serves the functions of various special features of the ATmega32A as listed on page 60.

2.2.7 RESETReset Input. A low level on this pin for longer than the minimum pulse length will generate a reset, even if the clockis not running. The minimum pulse length is given in Table 28-1 on page 281. Shorter pulses are not guaranteed togenerate a reset.

2.2.8 XTAL1Input to the inverting Oscillator amplifier and input to the internal clock operating circuit.

2.2.9 XTAL2Output from the inverting Oscillator amplifier.

2.2.10 AVCCAVCC is the supply voltage pin for Port A and the A/D Converter. It should be externally connected to VCC, even ifthe ADC is not used. If the ADC is used, it should be connected to VCC through a low-pass filter.

2.2.11 AREFAREF is the analog reference pin for the A/D Converter.

3. Resources A comprehensive set of development tools, application notes and datasheets are available for download onhttp://www.atmel.com/avr.

Note: 1.

4. Data RetentionReliability Qualification results show that the projected data retention failure rate is much less than 1 PPM over 20years at 85°C or 100 years at 25°C.


5. About Code Examples This documentation contains simple code examples that briefly show how to use various parts of the device. Thesecode examples assume that the part specific header file is included before compilation. Be aware that not all CCompiler vendors include bit definitions in the header files and interrupt handling in C is compiler dependent.Please confirm with the C Compiler documentation for more details.

6. Capacitive touch sensingThe Atmel® QTouch® Library provides a simple to use solution to realize touch sensitive interfaces on most AtmelAVR® microcontrollers. The QTouch Library includes support for the QTouch and QMatrix® acquisition methods.

Touch sensing can be added to any application by linking the appropriate Atmel QTouch Library for the AVR Micro-controller. This is done by using a simple set of APIs to define the touch channels and sensors, and then calling thetouch sensing API’s to retrieve the channel information and determine the touch sensor states.

The QTouch Library is FREE and downloadable from the Atmel website at the following location:www.atmel.com/qtouchlibrary. For implementation details and other information, refer to the Atmel QTouch LibraryUser Guide - also available for download from the Atmel website.


www.atmel.com/qtouchlibrary

http://www.atmel.com/dyn/resources/prod_documents/doc8207.pdf

7. AVR CPU Core

7.1 OverviewThis section discusses the Atmel®AVR® core architecture in general. The main function of the CPU core is toensure correct program execution. The CPU must therefore be able to access memories, perform calculations,control peripherals, and handle interrupts.

Figure 7-1. Block Diagram of the AVR MCU Architecture

In order to maximize performance and parallelism, the AVR uses a Harvard architecture – with separate memoriesand buses for program and data. Instructions in the program memory are executed with a single level pipelining.While one instruction is being executed, the next instruction is pre-fetched from the program memory. This conceptenables instructions to be executed in every clock cycle. The program memory is In-System ReprogrammableFlash memory.

The fast-access Register File contains 32 × 8-bit general purpose working registers with a single clock cycleaccess time. This allows single-cycle Arithmetic Logic Unit (ALU) operation. In a typical ALU operation, two oper-ands are output from the Register File, the operation is executed, and the result is stored back in the Register File– in one clock cycle.

Six of the 32 registers can be used as three 16-bit indirect address register pointers for Data Space addressing –enabling efficient address calculations. One of the these address pointers can also be used as an address pointerfor look up tables in Flash Program memory. These added function registers are the 16-bit X-, Y-, and Z-register,described later in this section.

FlashProgramMemory

InstructionRegister

InstructionDecoder

ProgramCounter

Control Lines

32 x 8GeneralPurpose

Registrers

ALU

Statusand Control

I/O Lines

EEPROM

Data Bus 8-bit

DataSRAM

Direct

Addre

ssin

g

Indirect

Addre

ssin

g

InterruptUnit

SPIUnit

WatchdogTimer

AnalogComparator

I/O Module 2

I/O Module1

I/O Module n


The ALU supports arithmetic and logic operations between registers or between a constant and a register. Singleregister operations can also be executed in the ALU. After an arithmetic operation, the Status Register is updatedto reflect information about the result of the operation.

Program flow is provided by conditional and unconditional jump and call instructions, able to directly address thewhole address space. Most AVR instructions have a single 16-bit word format. Every program memory addresscontains a 16-bit or 32-bit instruction.

Program Flash memory space is divided in two sections, the Boot program section and the Application Programsection. Both sections have dedicated Lock bits for write and read/write protection. The SPM instruction that writesinto the Application Flash memory section must reside in the Boot Program section.

During interrupts and subroutine calls, the return address Program Counter (PC) is stored on the Stack. The Stackis effectively allocated in the general data SRAM, and consequently the Stack size is only limited by the totalSRAM size and the usage of the SRAM. All user programs must initialize the SP in the reset routine (before sub-routines or interrupts are executed). The Stack Pointer SP is read/write accessible in the I/O space. The dataSRAM can easily be accessed through the five different addressing modes supported in the AVR architecture.

The memory spaces in the AVR architecture are all linear and regular memory maps.

A flexible interrupt module has its control registers in the I/O space with an additional global interrupt enable bit inthe Status Register. All interrupts have a separate interrupt vector in the interrupt vector table. The interrupts havepriority in accordance with their interrupt vector position. The lower the interrupt vector address, the higher thepriority.

The I/O memory space contains 64 addresses for CPU peripheral functions as Control Registers, SPI, and otherI/O functions. The I/O Memory can be accessed directly, or as the Data Space locations following those of the Reg-ister File, $20 - $5F.

7.2 ALU – Arithmetic Logic UnitThe high-performance AVR ALU operates in direct connection with all the 32 general purpose working registers.Within a single clock cycle, arithmetic operations between general purpose registers or between a register and animmediate are executed. The ALU operations are divided into three main categories – arithmetic, logical, and bit-functions. Some implementations of the architecture also provide a powerful multiplier supporting bothsigned/unsigned multiplication and fractional format. See the “Instruction Set” section for a detailed description.

7.3 Status RegisterThe Status Register contains information about the result of the most recently executed arithmetic instruction. Thisinformation can be used for altering program flow in order to perform conditional operations. Note that the StatusRegister is updated after all ALU operations, as specified in the Instruction Set Reference. This will in many casesremove the need for using the dedicated compare instructions, resulting in faster and more compact code.

The Status Register is not automatically stored when entering an interrupt routine and restored when returningfrom an interrupt. This must be handled by software.

7.3.1 SREG – AVR Status Register

Bit 7 6 5 4 3 2 1 0

I T H S V N Z C SREG

Read/Write R/W R/W R/W R/W R/W R/W R/W R/W

Initial Value 0 0 0 0 0 0 0 0


• Bit 7 – I: Global Interrupt EnableThe Global Interrupt Enable bit must be set for the interrupts to be enabled. The individual interrupt enable controlis then performed in separate control registers. If the Global Interrupt Enable Register is cleared, none of the inter-rupts are enabled independent of the individual interrupt enable settings. The I-bit is cleared by hardware after aninterrupt has occurred, and is set by the RETI instruction to enable subsequent interrupts. The I-bit can also be setand cleared by the application with the SEI and CLI instructions, as described in the instruction set reference.

• Bit 6 – T: Bit Copy StorageThe Bit Copy instructions BLD (Bit LoaD) and BST (Bit STore) use the T-bit as source or destination for the oper-ated bit. A bit from a register in the Register File can be copied into T by the BST instruction, and a bit in T can becopied into a bit in a register in the Register File by the BLD instruction.

• Bit 5 – H: Half Carry Flag The Half Carry Flag H indicates a half carry in some arithmetic operations. Half Carry is useful in BCD arithmetic.See the “Instruction Set Description” for detailed information.

• Bit 4 – S: Sign Bit, S = N VThe S-bit is always an exclusive or between the Negative Flag N and the Two’s Complement Overflow Flag V. Seethe “Instruction Set Description” for detailed information.

• Bit 3 – V: Two’s Complement Overflow FlagThe Two’s Complement Overflow Flag V supports two’s complement arithmetics. See the “Instruction Set Descrip-tion” for detailed information.

• Bit 2 – N: Negative FlagThe Negative Flag N indicates a negative result in an arithmetic or logic operation. See the “Instruction SetDescription” for detailed information.

• Bit 1 – Z: Zero FlagThe Zero Flag Z indicates a zero result in an arithmetic or logic operation. See the “Instruction Set Description” fordetailed information.

• Bit 0 – C: Carry FlagThe Carry Flag C indicates a carry in an arithmetic or logic operation. See the “Instruction Set Description” fordetailed information.

7.4 General Purpose Register FileThe Register File is optimized for the AVR Enhanced RISC instruction set. In order to achieve the required perfor-mance and flexibility, the following input/output schemes are supported by the Register File:

• One 8-bit output operand and one 8-bit result input

• Two 8-bit output operands and one 8-bit result input

• Two 8-bit output operands and one 16-bit result input

• One 16-bit output operand and one 16-bit result input

Figure 7-2 shows the structure of the 32 general purpose working registers in the CPU.


Figure 7-2. AVR CPU General Purpose Working Registers

Most of the instructions operating on the Register File have direct access to all registers, and most of them are sin-gle cycle instructions.

As shown in Figure 7-2, each register is also assigned a data memory address, mapping them directly into the first32 locations of the user Data Space. Although not being physically implemented as SRAM locations, this memoryorganization provides great flexibility in access of the registers, as the X-, Y-, and Z-pointer Registers can be set toindex any register in the file.

7 0 Addr.

R0 $00

R1 $01

R2 $02

…

R13 $0D

General R14 $0E

Purpose R15 $0F

Working R16 $10

Registers R17 $11

…

R26 $1A X-register Low Byte

R27 $1B X-register High Byte

R28 $1C Y-register Low Byte

R29 $1D Y-register High Byte

R30 $1E Z-register Low Byte

R31 $1F Z-register High Byte


7.4.1 The X-register, Y-register and Z-registerThe registers R26:R31 have some added functions to their general purpose usage. These registers are 16-bitaddress pointers for indirect addressing of the Data Space. The three indirect address registers X, Y, and Z aredefined as described in Figure 7-3.

Figure 7-3. The X-, Y-, and Z-registers

In the different addressing modes these address registers have functions as fixed displacement, automatic incre-ment, and automatic decrement (see the Instruction Set Reference for details).

7.5 Stack PointerThe Stack is mainly used for storing temporary data, for storing local variables and for storing return addressesafter interrupts and subroutine calls. Note that the Stack is implemented as growing from higher to lower memorylocations. The Stack Pointer Register always points to the top of the Stack. The Stack Pointer points to the dataSRAM Stack area where the Subroutine and Interrupt Stacks are located. A Stack PUSH command will decreasethe Stack Pointer.

The Stack in the data SRAM must be defined by the program before any subroutine calls are executed or interruptsare enabled. Initial Stack Pointer value equals the last address of the internal SRAM and the Stack Pointer must beset to point above start of the SRAM, see Figure 8-2 on page 16.

See Table 7-1 on page 11 for Stack Pointer details.

The AVR Stack Pointer is implemented as two 8-bit registers in the I/O space. The number of bits actually used isimplementation dependent. Note that the data space in some implementations of the AVR architecture is so smallthat only SPL is needed. In this case, the SPH Register will not be present.

15 XH XL 0

X - register 7 0 7 0

R27 ($1B) R26 ($1A)

15 YH YL 0

Y - register 7 0 7 0

R29 ($1D) R28 ($1C)

15 ZH ZL 0

Z - register 7 0 7 0

R31 ($1F) R30 ($1E)

Table 7-1. Stack Pointer instructions

Instruction Stack pointer Description

PUSH Decremented by 1 Data is pushed onto the stack

CALLICALLRCALL

Decremented by 2Return address is pushed onto the stack with a subroutine call or interrupt

POP Incremented by 1 Data is popped from the stack

RETRETI

Incremented by 2 Return address is popped from the stack with return from subroutine or return from interrupt


7.5.1 SPH and SPL – Stack Pointer High and Low Register

7.6 Instruction Execution TimingThis section describes the general access timing concepts for instruction execution. The AVR CPU is driven by theCPU clock clkCPU, directly generated from the selected clock source for the chip. No internal clock division is used.

Figure 7-4 shows the parallel instruction fetches and instruction executions enabled by the Harvard architectureand the fast-access Register File concept. This is the basic pipelining concept to obtain up to 1 MIPS per MHz withthe corresponding unique results for functions per cost, functions per clocks, and functions per power-unit.

Figure 7-4. The Parallel Instruction Fetches and Instruction Executions

Figure 7-5 shows the internal timing concept for the Register File. In a single clock cycle an ALU operation usingtwo register operands is executed, and the result is stored back to the destination register.

Figure 7-5. Single Cycle ALU Operation

Bit 15 14 13 12 11 10 9 8

SP15 SP14 SP13 SP12 SP11 SP10 SP9 SP8 SPH

SP7 SP6 SP5 SP4 SP3 SP2 SP1 SP0 SPL

7 6 5 4 3 2 1 0


R/W R/W R/W R/W R/W R/W R/W R/W


0 0 0 0 0 0 0 0

clk

1st Instruction Fetch

1st Instruction Execute2nd Instruction Fetch

2nd Instruction Execute3rd Instruction Fetch

3rd Instruction Execute4th Instruction Fetch

T1 T2 T3 T4

CPU

Total Execution Time

Register Operands Fetch

ALU Operation Execute

Result Write Back

T1 T2 T3 T4

clkCPU


7.7 Reset and Interrupt HandlingThe Atmel®AVR® provides several different interrupt sources. These interrupts and the separate reset vector eachhave a separate program vector in the program memory space. All interrupts are assigned individual enable bitswhich must be written logic one together with the Global Interrupt Enable bit in the Status Register in order toenable the interrupt. Depending on the Program Counter value, interrupts may be automatically disabled whenBoot Lock bits BLB02 or BLB12 are programmed. This feature improves software security. See the section “Mem-ory Programming” on page 248 for details.

The lowest addresses in the program memory space are by default defined as the Reset and Interrupt Vectors.The complete list of vectors is shown in “Interrupts” on page 43. The list also determines the priority levels of thedifferent interrupts. The lower the address the higher is the priority level. RESET has the highest priority, and nextis INT0 – the External Interrupt Request 0. The Interrupt Vectors can be moved to the start of the Boot Flash sec-tion by setting the IVSEL bit in the General Interrupt Control Register (GICR). Refer to “Interrupts” on page 43 formore information. The Reset Vector can also be moved to the start of the boot Flash section by programming theBOOTRST fuse, see “Boot Loader Support – Read-While-Write Self-Programming” on page 235.

When an interrupt occurs, the Global Interrupt Enable I-bit is cleared and all interrupts are disabled. The user soft-ware can write logic one to the I-bit to enable nested interrupts. All enabled interrupts can then interrupt the currentinterrupt routine. The I-bit is automatically set when a Return from Interrupt instruction – RETI – is executed.

There are basically two types of interrupts. The first type is triggered by an event that sets the Interrupt Flag. Forthese interrupts, the Program Counter is vectored to the actual Interrupt Vector in order to execute the interrupthandling routine, and hardware clears the corresponding Interrupt Flag. Interrupt Flags can also be cleared by writ-ing a logic one to the flag bit position(s) to be cleared. If an interrupt condition occurs while the correspondinginterrupt enable bit is cleared, the Interrupt Flag will be set and remembered until the interrupt is enabled, or theflag is cleared by software. Similarly, if one or more interrupt conditions occur while the Global Interrupt Enable bitis cleared, the corresponding Interrupt Flag(s) will be set and remembered until the global interrupt enable bit isset, and will then be executed by order of priority.

The second type of interrupts will trigger as long as the interrupt condition is present. These interrupts do not nec-essarily have Interrupt Flags. If the interrupt condition disappears before the interrupt is enabled, the interrupt willnot be triggered.

When the Atmel®AVR® exits from an interrupt, it will always return to the main program and execute one moreinstruction before any pending interrupt is served.

Note that the Status Register is not automatically stored when entering an interrupt routine, nor restored whenreturning from an interrupt routine. This must be handled by software.

When using the CLI instruction to disable interrupts, the interrupts will be immediately disabled. No interrupt will beexecuted after the CLI instruction, even if it occurs simultaneously with the CLI instruction. The following exampleshows how this can be used to avoid interrupts during the timed EEPROM write sequence.


When using the SEI instruction to enable interrupts, the instruction following SEI will be executed before any pend-ing interrupts, as shown in this example.

7.7.1 Interrupt Response TimeThe interrupt execution response for all the enabled AVR interrupts is four clock cycles minimum. After four clockcycles the program vector address for the actual interrupt handling routine is executed. During this four clock cycleperiod, the Program Counter is pushed onto the Stack. The vector is normally a jump to the interrupt routine, andthis jump takes three clock cycles. If an interrupt occurs during execution of a multi-cycle instruction, this instructionis completed before the interrupt is served. If an interrupt occurs when the MCU is in sleep mode, the interrupt exe-cution response time is increased by four clock cycles. This increase comes in addition to the start-up time from theselected sleep mode.

A return from an interrupt handling routine takes four clock cycles. During these four clock cycles, the ProgramCounter (two bytes) is popped back from the Stack, the Stack Pointer is incremented by two, and the I-bit in SREGis set.

Assembly Code Example

in r16, SREG ; store SREG value

cli ; disable interrupts during timed sequence

sbi EECR, EEMWE ; start EEPROM write

sbi EECR, EEWE

out SREG, r16 ; restore SREG value (I-bit)

C Code Example

char cSREG;

cSREG = SREG; /* store SREG value */

/* disable interrupts during timed sequence */

_CLI();

EECR |= (1<<EEMWE); /* start EEPROM write */

EECR |= (1<<EEWE);

SREG = cSREG; /* restore SREG value (I-bit) */


sei ; set global interrupt enable

sleep; enter sleep, waiting for interrupt

; note: will enter sleep before any pending

; interrupt(s)

C Code Example

_SEI(); /* set global interrupt enable */

_SLEEP(); /* enter sleep, waiting for interrupt */

/* note: will enter sleep before any pending interrupt(s) */


8. AVR Memories

8.1 OverviewThis section describes the different memories in the ATmega32A. The AVR architecture has two main memoryspaces, the Data Memory and the Program Memory space. In addition, the ATmega32A features an EEPROMMemory for data storage. All three memory spaces are linear and regular.

8.2 In-System Reprogrammable Flash Program Memory The ATmega32A contains 32Kbytes On-chip In-System Reprogrammable Flash memory for program storage.Since all AVR instructions are 16 bits or 32 bits wide, the Flash is organized as 16K × 16. For software security, theFlash Program memory space is divided into two sections, Boot Program section and Application Program section.

The Flash memory has an endurance of at least 10,000 write/erase cycles. The ATmega32A Program Counter(PC) is 14 bits wide, thus addressing the 16K program memory locations. The operation of Boot Program sectionand associated Boot Lock bits for software protection are described in detail in “Boot Loader Support – Read-While-Write Self-Programming” on page 235. “Memory Programming” on page 248 contains a detailed descriptionon Flash Programming in SPI, JTAG, or Parallel Programming mode.

Constant tables can be allocated within the entire program memory address space (see the LPM – Load ProgramMemory Instruction Description).

Timing diagrams for instruction fetch and execution are presented in “Instruction Execution Timing” on page 12.

Figure 8-1. Program Memory Map

$0000

$3FFF

Application Flash Section

Boot Flash Section


8.3 SRAM Data MemoryFigure 8-2 shows how the Atmel®AVR® ATmega32A SRAM Memory is organized.

The lower 2144 Data Memory locations address the Register File, the I/O Memory, and the internal data SRAM.The first 96 locations address the Register File and I/O Memory, and the next 2048 locations address the internaldata SRAM.

The five different addressing modes for the data memory cover: Direct, Indirect with Displacement, Indirect, Indi-rect with Pre-decrement, and Indirect with Post-increment. In the Register File, registers R26 to R31 feature theindirect Addressing Pointer Registers.

The direct addressing reaches the entire data space.

The Indirect with Displacement mode reaches 63 address locations from the base address given by the Y- or Z-register.

When using register indirect addressing modes with automatic pre-decrement and post-increment, the addressregisters X, Y, and Z are decremented or incremented.

The 32 general purpose working registers, 64 I/O Registers, and the 2048bytes of internal data SRAM in theATmega32A are all accessible through all these addressing modes. The Register File is described in “General Pur-pose Register File” on page 9.

Figure 8-2. Data Memory Map

8.3.1 Data Memory Access TimesThis section describes the general access timing concepts for internal memory access. The internal data SRAMaccess is performed in two clkCPU cycles as described in Figure 8-3.

Register File

R0R1R2

R29R30R31

I/O Registers$00$01$02

...

$3D$3E$3F

...

$0000$0001$0002

$001D$001E$001F

$0020$0021$0022

...

$005D$005E$005F

...

Data Address Space

$0060$0061

$085E$085F

...

Internal SRAM


Figure 8-3. On-chip Data SRAM Access Cycles

8.4 EEPROM Data MemoryThe Atmel®AVR® ATmega32A contains 1024bytes of data EEPROM memory. It is organized as a separate dataspace, in which single bytes can be read and written. The EEPROM has an endurance of at least 100,000write/erase cycles. The access between the EEPROM and the CPU is described in the following, specifying theEEPROM Address Registers, the EEPROM Data Register, and the EEPROM Control Register.

“Memory Programming” on page 248 contains a detailed description on EEPROM Programming in SPI, JTAG, orParallell Programming mode.

8.4.1 EEPROM Read/Write AccessThe EEPROM Access Registers are accessible in the I/O space.

The write access time for the EEPROM is given in Table 8-1 on page 20. A self-timing function, however, lets theuser software detect when the next byte can be written. If the user code contains instructions that write theEEPROM, some precautions must be taken. In heavily filtered power supplies, VCC is likely to rise or fall slowly onPower-up/down. This causes the device for some period of time to run at a voltage lower than specified as mini-mum for the clock frequency used. See “Preventing EEPROM Corruption” on page 18 for details on how to avoidproblems in these situations.

In order to prevent unintentional EEPROM writes, a specific write procedure must be followed. Refer to thedescription of the EEPROM Control Register for details on this.

When the EEPROM is read, the CPU is halted for four clock cycles before the next instruction is executed. Whenthe EEPROM is written, the CPU is halted for two clock cycles before the next instruction is executed.

8.4.2 EEPROM Write During Power-down Sleep ModeWhen entering Power-down Sleep mode while an EEPROM write operation is active, the EEPROM write operationwill continue, and will complete before the Write Access time has passed. However, when the write operation iscompleted, the Oscillator continues running, and as a consequence, the device does not enter Power-downentirely. It is therefore recommended to verify that the EEPROM write operation is completed before enteringPower-down.

clk

WR

RD

Data

Data

Address Address Valid

T1 T2 T3

Compute Address

Rea

dW

rite

CPU

Memory Access Instruction Next Instruction


8.4.3 Preventing EEPROM CorruptionDuring periods of low VCC, the EEPROM data can be corrupted because the supply voltage is too low for the CPUand the EEPROM to operate properly. These issues are the same as for board level systems using EEPROM, andthe same design solutions should be applied.

An EEPROM data corruption can be caused by two situations when the voltage is too low. First, a regular writesequence to the EEPROM requires a minimum voltage to operate correctly. Secondly, the CPU itself can executeinstructions incorrectly, if the supply voltage is too low.

EPROM data corruption can easily be avoided by following this design recommendation:

Keep the AVR RESET active (low) during periods of insufficient power supply voltage. This can be done byenabling the internal Brown-out Detector (BOD). If the detection level of the internal BOD does not match theneeded detection level, an external low VCC Reset Protection circuit can be used. If a reset occurs while a writeoperation is in progress, the write operation will be completed provided that the power supply voltage issufficient.

8.5 I/O MemoryThe I/O space definition of the Atmel®AVR® ATmega32A is shown in “Register Summary” on page 316.

All ATmega32A I/Os and peripherals are placed in the I/O space. The I/O locations are accessed by the IN andOUT instructions, transferring data between the 32 general purpose working registers and the I/O space. I/O Reg-isters within the address range $00 - $1F are directly bit-accessible using the SBI and CBI instructions. In theseregisters, the value of single bits can be checked by using the SBIS and SBIC instructions. Refer to the InstructionSet section for more details. When using the I/O specific commands IN and OUT, the I/O addresses $00 - $3Fmust be used. When addressing I/O Registers as data space using LD and ST instructions, $20 must be added tothese addresses.

For compatibility with future devices, reserved bits should be written to zero if accessed. Reserved I/O memoryaddresses should never be written.

Some of the Status Flags are cleared by writing a logical one to them. Note that the CBI and SBI instructions willoperate on all bits in the I/O Register, writing a one back into any flag read as set, thus clearing the flag. The CBIand SBI instructions work with registers $00 to $1F only.

The I/O and Peripherals Control Registers are explained in later sections.


8.6 Register Description

8.6.1 EEARH and EEARL – EEPROM Address Register

• Bits [15:10] – Reserved BitsThese bits are reserved bits in the ATmega32A and will always read as zero.

• Bits [9:0] – EEAR9:0: EEPROM AddressThe EEPROM Address Registers – EEARH and EEARL – specify the EEPROM address in the 1024bytesEEPROM space. The EEPROM data bytes are addressed linearly between 0 and 1023. The initial value of EEARis undefined. A proper value must be written before the EEPROM may be accessed.

8.6.2 EEDR – EEPROM Data Register

• Bits [7:0] – EEDR7.0: EEPROM DataFor the EEPROM write operation, the EEDR Register contains the data to be written to the EEPROM in theaddress given by the EEAR Register. For the EEPROM read operation, the EEDR contains the data read out fromthe EEPROM at the address given by EEAR.

8.6.3 EECR – EEPROM Control Register

• Bits [7:4] – Reserved BitsThese bits are reserved bits in the ATmega32A and will always read as zero.

• Bit 3 – EERIE: EEPROM Ready Interrupt EnableWriting EERIE to one enables the EEPROM Ready Interrupt if the I bit in SREG is set. Writing EERIE to zero dis-ables the interrupt. The EEPROM Ready interrupt generates a constant interrupt when EEWE is cleared.

• Bit 2 – EEMWE: EEPROM Master Write EnableThe EEMWE bit determines whether setting EEWE to one causes the EEPROM to be written. When EEMWE isset, setting EEWE within four clock cycles will write data to the EEPROM at the selected address If EEMWE iszero, setting EEWE will have no effect. When EEMWE has been written to one by software, hardware clears the bitto zero after four clock cycles. See the description of the EEWE bit for an EEPROM write procedure.

Bit 15 14 13 12 11 10 9 8

– – – – – – EEAR9 EEAR8 EEARH

EEAR7 EEAR6 EEAR5 EEAR4 EEAR3 EEAR2 EEAR1 EEAR0 EEARL

7 6 5 4 3 2 1 0

Read/Write R R R R R R R/W R/W


Initial Value 0 0 0 0 0 0 0 X

X X X X X X X X

Bit 7 6 5 4 3 2 1 0

MSB LSB EEDR



Bit 7 6 5 4 3 2 1 0

– – – – EERIE EEMWE EEWE EERE EECR

Read/Write R R R R R/W R/W R/W R/W

Initial Value 0 0 0 0 0 0 X 0


• Bit 1 – EEWE: EEPROM Write EnableThe EEPROM Write Enable Signal EEWE is the write strobe to the EEPROM. When address and data are cor-rectly set up, the EEWE bit must be written to one to write the value into the EEPROM. The EEMWE bit must bewritten to one before a logical one is written to EEWE, otherwise no EEPROM write takes place. The following pro-cedure should be followed when writing the EEPROM (the order of steps 3 and 4 is not essential):

1. Wait until EEWE becomes zero.

2. Wait until SPMEN in SPMCR becomes zero.

3. Write new EEPROM address to EEAR (optional).

4. Write new EEPROM data to EEDR (optional).

5. Write a logical one to the EEMWE bit while writing a zero to EEWE in EECR.

6. Within four clock cycles after setting EEMWE, write a logical one to EEWE.

The EEPROM can not be programmed during a CPU write to the Flash memory. The software must check that theFlash programming is completed before initiating a new EEPROM write. Step 2 is only relevant if the software con-tains a Boot Loader allowing the CPU to program the Flash. If the Flash is never being updated by the CPU, step 2can be omitted. See “Boot Loader Support – Read-While-Write Self-Programming” on page 235 for details aboutboot programming.

Caution: An interrupt between step 5 and step 6 will make the write cycle fail, since the EEPROM Master WriteEnable will time-out. If an interrupt routine accessing the EEPROM is interrupting another EEPROM Access, theEEAR or EEDR reGister will be modified, causing the interrupted EEPROM Access to fail. It is recommended tohave the Global Interrupt Flag cleared during all the steps to avoid these problems.

When the write access time has elapsed, the EEWE bit is cleared by hardware. The user software can poll this bitand wait for a zero before writing the next byte. When EEWE has been set, the CPU is halted for two cycles beforethe next instruction is executed.

• Bit 0 – EERE: EEPROM Read EnableThe EEPROM Read Enable Signal – EERE – is the read strobe to the EEPROM. When the correct address is setup in the EEAR Register, the EERE bit must be written to a logic one to trigger the EEPROM read. The EEPROMread access takes one instruction, and the requested data is available immediately. When the EEPROM is read,the CPU is halted for four cycles before the next instruction is executed.

The user should poll the EEWE bit before starting the read operation. If a write operation is in progress, it is neitherpossible to read the EEPROM, nor to change the EEAR Register.

The calibrated Oscillator is used to time the EEPROM accesses. Table 8-1 lists the typical programming time forEEPROM access from the CPU.

Note: 1. Uses 1MHz clock, independent of CKSEL Fuse setting.

The following code examples show one assembly and one C function for writing to the EEPROM. The examplesassume that interrupts are controlled (for example by disabling interrupts globally) so that no interrupts will occurduring execution of these functions. The examples also assume that no Flash Boot Loader is present in the soft-ware. If such code is present, the EEPROM write function must also wait for any ongoing SPM command to finish.

Table 8-1. EEPROM Programming Time

SymbolNumber of Calibrated RC Oscillator

Cycles(1) Typ Programming Time

EEPROM write (from CPU) 8448 8.5ms



EEPROM_write:

; Wait for completion of previous write

sbic EECR,EEWE

rjmp EEPROM_write

; Set up address (r18:r17) in address register

out EEARH, r18

out EEARL, r17

; Write data (r16) to data register

out EEDR,r16

; Write logical one to EEMWE

sbi EECR,EEMWE

; Start eeprom write by setting EEWE

sbi EECR,EEWE

ret

C Code Example

void EEPROM_write(unsigned int uiAddress, unsigned char ucData)

{

/* Wait for completion of previous write */

while(EECR & (1<<EEWE))

;

/* Set up address and data registers */

EEAR = uiAddress;

EEDR = ucData;

/* Write logical one to EEMWE */

EECR |= (1<<EEMWE);

/* Start eeprom write by setting EEWE */

EECR |= (1<<EEWE);

}


The next code examples show assembly and C functions for reading the EEPROM. The examples assume thatinterrupts are controlled so that no interrupts will occur during execution of these functions.


EEPROM_read:

; Wait for completion of previous write

sbic EECR,EEWE

rjmp EEPROM_read

; Set up address (r18:r17) in address register

out EEARH, r18

out EEARL, r17

; Start eeprom read by writing EERE

sbi EECR,EERE

; Read data from data register

in r16,EEDR

ret

C Code Example

unsigned char EEPROM_read(unsigned int uiAddress)

{

/* Wait for completion of previous write */

while(EECR & (1<<EEWE))

;

/* Set up address register */

EEAR = uiAddress;

/* Start eeprom read by writing EERE */

EECR |= (1<<EERE);

/* Return data from data register */

return EEDR;

}


9. System Clock and Clock Options

9.1 Clock Systems and their DistributionFigure 9-1 presents the principal clock systems in the AVR and their distribution. All of the clocks need not beactive at a given time. In order to reduce power consumption, the clocks to modules not being used can be haltedby using different sleep modes, as described in ““Power Management and Sleep Modes” on page 31. The clocksystems are detailed Figure 9-1.

Figure 9-1. Clock Distribution

9.1.1 CPU Clock – clkCPU

The CPU clock is routed to parts of the system concerned with operation of the AVR core. Examples of such mod-ules are the General Purpose Register File, the Status Register and the data memory holding the Stack Pointer.Halting the CPU clock inhibits the core from performing general operations and calculations.

9.1.2 I/O Clock – clkI/O

The I/O clock is used by the majority of the I/O modules, like Timer/Counters, SPI, and USART. The I/O clock isalso used by the External Interrupt module, but note that some external interrupts are detected by asynchronouslogic, allowing such interrupts to be detected even if the I/O clock is halted. Also note that address recognition inthe TWI module is carried out asynchronously when clkI/O is halted, enabling TWI address reception in all sleepmodes.

9.1.3 Flash Clock – clkFLASH

The Flash clock controls operation of the Flash interface. The Flash clock is usually active simultaneously with theCPU clock.

General I/OModules

AsynchronousTimer/Counter

ADC CPU Core RAM

clkI/O

clkASY

AVR ClockControl Unit

clkCPU

Flash andEEPROM

clkFLASH

clkADC

Source Clock

Watchdog Timer

WatchdogOscillator

Reset Logic

ClockMultiplexer

Watchdog Clock

Calibrated RCOscillator

Timer/CounterOscillator

CrystalOscillator

Low-frequencyCrystal Oscillator

External RCOscillator

External Clock


9.1.4 Asynchronous Timer Clock – clkASY

The Asynchronous Timer clock allows the Asynchronous Timer/Counter to be clocked directly from an external32kHz clock crystal. The dedicated clock domain allows using this Timer/Counter as a real-time counter even whenthe device is in sleep mode.

9.1.5 ADC Clock – clkADC

The ADC is provided with a dedicated clock domain. This allows halting the CPU and I/O clocks in order to reducenoise generated by digital circuitry. This gives more accurate ADC conversion results.

9.2 Clock SourcesThe device has the following clock source options, selectable by Flash Fuse bits as shown below. The clock fromthe selected source is input to the AVR clock generator, and routed to the appropriate modules.

Note: 1. For all fuses “1” means unprogrammed while “0” means programmed.

The various choices for each clocking option is given in the following sections. When the CPU wakes up fromPower-down or Power-save, the selected clock source is used to time the start-up, ensuring stable Oscillator oper-ation before instruction execution starts. When the CPU starts from Reset, there is as an additional delay allowingthe power to reach a stable level before commencing normal operation. The Watchdog Oscillator is used for timingthis real-time part of the start-up time. The number of WDT Oscillator cycles used for each time-out is shown inTable 9-2. The frequency of the Watchdog Oscillator is voltage dependent as shown in “Register Summary” onpage 316.

9.3 Default Clock Source The device is shipped with CKSEL = “0001” and SUT = “10”. The default clock source setting is therefore the 1MHzInternal RC Oscillator with longest startup time. This default setting ensures that all users can make their desiredclock source setting using an In-System or Parallel Programmer.

9.4 Crystal OscillatorXTAL1 and XTAL2 are input and output, respectively, of an inverting amplifier which can be configured for use asan On-chip Oscillator, as shown in Figure 9-2. Either a quartz crystal or a ceramic resonator may be used. TheCKOPT Fuse selects between two different Oscillator amplifier modes. When CKOPT is programmed, the Oscilla-tor output will oscillate will a full rail-to-rail swing on the output. This mode is suitable when operating in a very noisyenvironment or when the output from XTAL2 drives a second clock buffer. This mode has a wide frequency range.When CKOPT is unprogrammed, the Oscillator has a smaller output swing. This reduces power consumption con-siderably. This mode has a limited frequency range and it can not be used to drive other clock buffers.

Table 9-1. Device Clocking Options Select(1)

Device Clocking Option CKSEL3:0

External Crystal/Ceramic Resonator 1111 - 1010

External Low-frequency Crystal 1001

External RC Oscillator 1000 - 0101

Calibrated Internal RC Oscillator 0100 - 0001

External Clock 0000

Table 9-2. Number of Watchdog Oscillator Cycles

Typ Time-out (VCC = 5.0V) Typ Time-out (VCC = 3.0V) Number of Cycles

4.1ms 4.3ms 4K (4,096)

65ms 69ms 64K (65,536)


For resonators, the maximum frequency is 8MHz with CKOPT unprogrammed and 16 MHz with CKOPT pro-grammed. C1 and C2 should always be equal for both crystals and resonators. The optimal value of the capacitorsdepends on the crystal or resonator in use, the amount of stray capacitance, and the electromagnetic noise of theenvironment. Some initial guidelines for choosing capacitors for use with crystals are given in Table 9-3. Forceramic resonators, the capacitor values given by the manufacturer should be used.

Figure 9-2. Crystal Oscillator Connections

The Oscillator can operate in three different modes, each optimized for a specific frequency range. The operatingmode is selected by the fuses CKSEL3:1 as shown in Table 9-3.

Note: 1. This option should not be used with crystals, only with ceramic resonators.

The CKSEL0 Fuse together with the SUT1:0 fuses select the start-up times as shown in Table 9-4.

Table 9-3. Crystal Oscillator Operating Modes

CKOPT CKSEL3:1 Frequency Range

(MHz)Recommended Range for Capacitors C1

and C2 for Use with Crystals (pF)

1 101(1) 0.4 - 0.9 –

1 110 0.9 - 3.0 12 - 22

1 111 3.0 - 8.0 12 - 22

0 101, 110, 111 1.0 12 - 22

Table 9-4. Start-up Times for the Crystal Oscillator Clock Selection

CKSEL0 SUT1:0

Start-up Time fromPower-down and

Power-save

Additional Delayfrom Reset(VCC = 5.0V) Recommended Usage

0 00 258 CK(1) 4.1msCeramic resonator, fast rising power

0 01 258 CK(1) 65msCeramic resonator, slowly rising power

0 10 1K CK(2) –Ceramic resonator, BOD enabled

0 11 1K CK(2) 4.1msCeramic resonator, fast rising power

1 00 1K CK(2) 65msCeramic resonator, slowly rising power

XTAL2

XTAL1

GND

C2

C1


Notes: 1. These options should only be used when not operating close to the maximum frequency of the device, and only if frequency stability at start-up is not important for the application. These options are not suitable for crystals.

2. These options are intended for use with ceramic resonators and will ensure frequency stability at start-up. They can also be used with crystals when not operating close to the maximum frequency of the device, and if frequency sta-bility at start-up is not important for the application.

9.5 Low-frequency Crystal OscillatorTo use a 32.768kHz watch crystal as the clock source for the device, the Low-frequency Crystal Oscillator must beselected by setting the CKSEL fuses to “1001”. The crystal should be connected as shown in Figure 9-2. By pro-gramming the CKOPT Fuse, the user can enable internal capacitors on XTAL1 and XTAL2, thereby removing theneed for external capacitors. The internal capacitors have a nominal value of 36 pF.

When this Oscillator is selected, start-up times are determined by the SUT fuses as shown in Table 9-5.

Note: 1. These options should only be used if frequency stability at start-up is not important for the application.

9.6 External RC OscillatorFor timing insensitive applications, the external RC configuration shown in Figure 9-3 can be used. The frequencyis roughly estimated by the equation f = 1/(3RC). C should be at least 22 pF. By programming the CKOPT Fuse,the user can enable an internal 36 pF capacitor between XTAL1 and GND, thereby removing the need for an exter-nal capacitor. For more information on Oscillator operation and details on how to choose R and C, refer to theExternal RC Oscillator application note.

1 01 16K CK –Crystal Oscillator, BOD enabled

1 10 16K CK 4.1msCrystal Oscillator, fast rising power

1 11 16K CK 65msCrystal Oscillator, slowly rising power

Table 9-4. Start-up Times for the Crystal Oscillator Clock Selection (Continued)

CKSEL0 SUT1:0


Power-save

Additional Delayfrom Reset(VCC = 5.0V) Recommended Usage

Table 9-5. Start-up Times for the Low-frequency Crystal Oscillator Clock Selection

SUT1:0


Power-save

Additional Delay from Reset (VCC = 5.0V) Recommended Usage

00 1K CK(1) 4.1ms Fast rising power or BOD enabled

01 1K CK(1) 65ms Slowly rising power

10 32K CK 65ms Stable frequency at start-up

11 Reserved


Figure 9-3. External RC Configuration

The Oscillator can operate in four different modes, each optimized for a specific frequency range. The operatingmode is selected by the fuses CKSEL3:0 as shown in Table 9-6.

When this Oscillator is selected, start-up times are determined by the SUT fuses as shown in Table 9-7.

Note: 1. This option should not be used when operating close to the maximum frequency of the device.

9.7 Calibrated Internal RC OscillatorThe Calibrated Internal RC Oscillator provides a fixed 1.0, 2.0, 4.0, or 8.0 MHz clock. All frequencies are nominalvalues at 5V and 25C. This clock may be selected as the system clock by programming the CKSEL fuses asshown in Table 9-8. If selected, it will operate with no external components. The CKOPT Fuse should always beunprogrammed when using this clock option. During Reset, hardware loads the calibration byte for the 1MHz intothe OSCCAL Register and thereby automatically calibrates the RC Oscillator. At 5V, 25C and 1.0MHz Oscillatorfrequency selected, this calibration gives a frequency within ±3% of the nominal frequency. Using calibration meth-ods as described in application notes available at www.atmel.com/avr it is possible to achieve ±1% accuracy at anygiven VCC and Temperature. When this Oscillator is used as the Chip Clock, the Watchdog Oscillator will still beused for the Watchdog Timer and for the reset time-out. For more information on the pre-programmed calibrationvalue, see the section “Calibration Byte” on page 250.

Table 9-6. External RC Oscillator Operating Modes

CKSEL3:0 Frequency Range (MHz)

0101 0.1 - 0.9

0110 0.9 - 3.0

0111 3.0 - 8.0

1000 8.0 - 12.0

Table 9-7. Start-up Times for the External RC Oscillator Clock Selection

SUT1:0


Power-save

Additional Delayfrom Reset (VCC = 5.0V) Recommended Usage

00 18 CK – BOD enabled

01 18 CK 4.1ms Fast rising power

10 18 CK 65ms Slowly rising power

11 6 CK(1) 4.1ms Fast rising power or BOD enabled

XTAL2

XTAL1

GNDC

R

VCC

NC


Note: 1. The device is shipped with this option selected.

When this Oscillator is selected, start-up times are determined by the SUT fuses as shown in Table 9-9. XTAL1and XTAL2 should be left unconnected (NC).

Note: 1. The device is shipped with this option selected.

9.8 External ClockTo drive the device from an external clock source, XTAL1 should be driven as shown in Figure 9-4 on page 28. Torun the device on an external clock, the CKSEL fuses must be programmed to “0000”. By programming theCKOPT Fuse, the user can enable an internal 36pF capacitor between XTAL1 and GND.

Figure 9-4. External Clock Drive Configuration

Table 9-8. Internal Calibrated RC Oscillator Operating Modes

CKSEL3:0 Nominal Frequency (MHz)

0001(1) 1.0

0010 2.0

0011 4.0

0100 8.0

Table 9-9. Start-up Times for the Internal Calibrated RC Oscillator Clock Selection

SUT1:0


Power-save




10(1) 6 CK 65ms Slowly rising power

11 Reserved

EXTERNALCLOCKSIGNAL


When this clock source is selected, start-up times are determined by the SUT fuses as shown in Table 9-10.

When applying an external clock, it is required to avoid sudden changes in the applied clock frequency to ensurestable operation of the MCU. A variation in frequency of more than 2% from one clock cycle to the next can lead tounpredictable behavior. It is required to ensure that the MCU is kept in reset during such changes in the clockfrequency.

9.9 Timer/Counter OscillatorFor AVR microcontrollers with Timer/Counter Oscillator pins (TOSC1 and TOSC2), the crystal is connected directlybetween the pins. No external capacitors are needed. The Oscillator is optimized for use with a 32.768kHz watchcrystal. Applying an external clock source to TOSC1 is not recommended.

Note: The Timer/Counter Oscillator uses the same type of crystal oscillator as Low-Frequency Oscillator and the internal capacitors have the same nominal value of 36pF.

Table 9-10. Start-up Times for the External Clock Selection

SUT1:0

Start-up Time from Power-down and

Power-save




10 6 CK 65ms Slowly rising power

11 Reserved



9.10.1 OSCCAL – Oscillator Calibration Register

• Bits 7:0 – CAL7:0: Oscillator Calibration ValueWriting the calibration byte to this address will trim the Internal Oscillator to remove process variations from theOscillator frequency. During Reset, the 1MHz calibration value which is located in the signature row High Byte(address 0x00) is automatically loaded into the OSCCAL Register. If the internal RC is used at other frequencies,the calibration values must be loaded manually. This can be done by first reading the signature row by a program-mer, and then store the calibration values in the Flash or EEPROM. Then the value can be read by software andloaded into the OSCCAL Register. When OSCCAL is zero, the lowest available frequency is chosen. Writing non-zero values to this register will increase the frequency of the Internal Oscillator. Writing $FF to the register givesthe highest available frequency. The calibrated Oscillator is used to time EEPROM and Flash access. If EEPROMor Flash is written, do not calibrate to more than 10% above the nominal frequency. Otherwise, the EEPROM orFlash write may fail. Note that the Oscillator is intended for calibration to 1.0, 2.0z, 4.0, or 8.0MHz. Tuning to othervalues is not guaranteed, as indicated in Table 9-11.

Bit 7 6 5 4 3 2 1 0

CAL7 CAL6 CAL5 CAL4 CAL3 CAL2 CAL1 CAL0 OSCCAL


Initial Value Device Specific Calibration Value

Table 9-11. Internal RC Oscillator Frequency Range.

OSCCAL ValueMin Frequency in Percentage of

Nominal Frequency (%)Max Frequency in Percentage of

Nominal Frequency (%)

$00 50 100

$7F 75 150

$FF 100 200


10. Power Management and Sleep Modes

10.1 Sleep ModesSleep modes enable the application to shut down unused modules in the MCU, thereby saving power. The AVRprovides various sleep modes allowing the user to tailor the power consumption to the application’s requirements.

Figure 8-1 on page 25 presents the different clock systems in the ATmega32A, and their distribution. The figure ishelpful in selecting an appropriate sleep mode. Table 10-1 shows the different clock options and their wake-upsources.

Notes: 1. External Crystal or resonator selected as clock source.

2. If AS2 bit in ASSR is set.

3. Only INT2 or level interrupt INT1 and INT0.

To enter any of the six sleep modes, the SE bit in MCUCR must be written to logic one and a SLEEP instructionmust be executed. The SM2, SM1, and SM0 bits in the MCUCR Register select which sleep mode (Idle, ADCNoise Reduction, Power-down, Power-save, Standby, or Extended Standby) will be activated by the SLEEPinstruction. See Table 10-2 on page 35 for a summary.

If an enabled interrupt occurs while the MCU is in a sleep mode, the MCU wakes up. The MCU is then halted forfour cycles in addition to the start-up time, it executes the interrupt routine, and resumes execution from the instruc-tion following SLEEP. The contents of the Register File and SRAM are unaltered when the device wakes up fromsleep. If a Reset occurs during sleep mode, the MCU wakes up and executes from the Reset Vector.

10.2 Idle ModeWhen the SM2:0 bits are written to 000, the SLEEP instruction makes the MCU enter Idle mode, stopping the CPUbut allowing SPI, USART, Analog Comparator, ADC, Two-wire Serial Interface, Timer/Counters, Watchdog, andthe interrupt system to continue operating. This sleep mode basically halts clkCPU and clkFLASH, while allowing theother clocks to run.

Idle mode enables the MCU to wake up from external triggered interrupts as well as internal ones like the TimerOverflow and USART Transmit Complete interrupts. If wake-up from the Analog Comparator interrupt is not

Table 10-1. Active Clock Domains and Wake Up Sources in the Different Sleep Modes

Active Clock domains Oscillators Wake-up Sources

Sleep Mode clk C

PU

clk F

LA

SH

clk I

O

clk A

DC

clk A

SY

Mai

n C

lock

S

ou

rce

En

able

d

Tim

er O

scill

ato

r E

nab

led

INT

2IN

T1

INT

0

TW

I Ad

dre

ss

Mat

ch

Tim

er2 S

PM

/ E

EP

RO

M

Rea

dy

AD

C

Oth

erI/O

Idle X X X X X(2) X X X X X X

ADC NoiseReduction

X X X X(2) X(3) X X X X

Power-down X(3) X

Power-save X(2) X(2) X(3) X X(2)

Standby(1) X X(3) X

ExtendedStandby(1) X(2) X X(2) X(3) X X(2)


required, the Analog Comparator can be powered down by setting the ACD bit in the Analog Comparator Controland Status Register – ACSR. This will reduce power consumption in Idle mode. If the ADC is enabled, a conver-sion starts automatically when this mode is entered.

10.3 ADC Noise Reduction ModeWhen the SM2:0 bits are written to 001, the SLEEP instruction makes the MCU enter ADC Noise Reduction mode,stopping the CPU but allowing the ADC, the External Interrupts, the Two-wire Serial Interface address watch,Timer/Counter2 and the Watchdog to continue operating (if enabled). This sleep mode basically halts clkI/O, clkCPU,and clkFLASH, while allowing the other clocks to run.

This improves the noise environment for the ADC, enabling higher resolution measurements. If the ADC isenabled, a conversion starts automatically when this mode is entered. Apart form the ADC Conversion Completeinterrupt, only an External Reset, a Watchdog Reset, a Brown-out Reset, a Two-wire Serial Interface AddressMatch Interrupt, a Timer/Counter2 interrupt, an SPM/EEPROM ready interrupt, an External level interrupt on INT0or INT1, or an external interrupt on INT2 can wake up the MCU from ADC Noise Reduction mode.

10.4 Power-down ModeWhen the SM2:0 bits are written to 010, the SLEEP instruction makes the MCU enter Power-down mode. In thismode, the External Oscillator is stopped, while the External interrupts, the Two-wire Serial Interface addresswatch, and the Watchdog continue operating (if enabled). Only an External Reset, a Watchdog Reset, a Brown-outReset, a Two-wire Serial Interface address match interrupt, an External level interrupt on INT0 or INT1, or anExternal interrupt on INT2 can wake up the MCU. This sleep mode basically halts all generated clocks, allowingoperation of asynchronous modules only.

Note that if a level triggered interrupt is used for wake-up from Power-down mode, the changed level must be heldfor some time to wake up the MCU. Refer to “External Interrupts” on page 65 for details.

When waking up from Power-down mode, there is a delay from the wake-up condition occurs until the wake-upbecomes effective. This allows the clock to restart and become stable after having been stopped. The wake-upperiod is defined by the same CKSEL fuses that define the reset time-out period, as described in “Clock Sources”on page 24.

10.5 Power-save ModeWhen the SM2:0 bits are written to 011, the SLEEP instruction makes the MCU enter Power-save mode. Thismode is identical to Power-down, with one exception:

If Timer/Counter2 is clocked asynchronously, that is, the AS2 bit in ASSR is set, Timer/Counter2 will run duringsleep. The device can wake up from either Timer Overflow or Output Compare event from Timer/Counter2 if thecorresponding Timer/Counter2 interrupt enable bits are set in TIMSK, and the Global Interrupt Enable bit in SREGis set.

If the Asynchronous Timer is NOT clocked asynchronously, Power-down mode is recommended instead of Power-save mode because the contents of the registers in the Asynchronous Timer should be considered undefined afterwake-up in Power-save mode if AS2 is 0.

This sleep mode basically halts all clocks except clkASY, allowing operation only of asynchronous modules, includ-ing Timer/Counter2 if clocked asynchronously.

10.6 Standby ModeWhen the SM2:0 bits are 110 and an external crystal/resonator clock option is selected, the SLEEP instructionmakes the MCU enter Standby mode. This mode is identical to Power-down with the exception that the Oscillator iskept running. From Standby mode, the device wakes up in six clock cycles.


10.7 Extended Standby ModeWhen the SM2:0 bits are 111 and an external crystal/resonator clock option is selected, the SLEEP instructionmakes the MCU enter Extended Standby mode. This mode is identical to Power-save mode with the exception thatthe Oscillator is kept running. From Extended Standby mode, the device wakes up in six clock cycles.

10.8 Minimizing Power ConsumptionThere are several issues to consider when trying to minimize the power consumption in an AVR controlled system.In general, sleep modes should be used as much as possible, and the sleep mode should be selected so that asfew as possible of the device’s functions are operating. All functions not needed should be disabled. In particular,the following modules may need special consideration when trying to achieve the lowest possible powerconsumption.

10.8.1 Analog to Digital ConverterIf enabled, the ADC will be enabled in all sleep modes. To save power, the ADC should be disabled before enteringany sleep mode. When the ADC is turned off and on again, the next conversion will be an extended conversion.Refer to “Analog to Digital Converter” on page 193 for details on ADC operation.

10.8.2 Analog ComparatorWhen entering Idle mode, the Analog Comparator should be disabled if not used. When entering ADC NoiseReduction mode, the Analog Comparator should be disabled. In the other sleep modes, the Analog Comparator isautomatically disabled. However, if the Analog Comparator is set up to use the Internal Voltage Reference asinput, the Analog Comparator should be disabled in all sleep modes. Otherwise, the Internal Voltage Reference willbe enabled, independent of sleep mode. Refer to “Analog Comparator” on page 190 for details on how to configurethe Analog Comparator.

10.8.3 Brown-out DetectorIf the Brown-out Detector is not needed in the application, this module should be turned off. If the Brown-out Detec-tor is enabled by the BODEN Fuse, it will be enabled in all sleep modes, and hence, always consume power. In thedeeper sleep modes, this will contribute significantly to the total current consumption. Refer to “Brown-out Detec-tor” on page 33 for details on how to configure the Brown-out Detector.

10.8.4 Internal Voltage ReferenceThe Internal Voltage Reference will be enabled when needed by the Brown-out Detector, the Analog Comparatoror the ADC. If these modules are disabled as described in the sections above, the internal voltage reference will bedisabled and it will not be consuming power. When turned on again, the user must allow the reference to start upbefore the output is used. If the reference is kept on in sleep mode, the output can be used immediately. Refer to“Internal Voltage Reference” on page 33 for details on the start-up time.

10.8.5 Watchdog TimerIf the Watchdog Timer is not needed in the application, this module should be turned off. If the Watchdog Timer isenabled, it will be enabled in all sleep modes, and hence, always consume power. In the deeper sleep modes, thiswill contribute significantly to the total current consumption. Refer to “Watchdog Timer” on page 33 for details onhow to configure the Watchdog Timer.

10.8.6 Port PinsWhen entering a sleep mode, all port pins should be configured to use minimum power. The most important thingis then to ensure that no pins drive resistive loads. In sleep modes where the both the I/O clock (clkI/O) and theADC clock (clkADC) are stopped, the input buffers of the device will be disabled. This ensures that no power is con-sumed by the input logic when not needed. In some cases, the input logic is needed for detecting wake-upconditions, and it will then be enabled. Refer to the section “Digital Input Enable and Sleep Modes” on page 52 for


details on which pins are enabled. If the input buffer is enabled and the input signal is left floating or have an analogsignal level close to VCC/2, the input buffer will use excessive power.

10.8.7 JTAG Interface and On-chip Debug System

• If the On-chip debug system is enabled by the OCDEN Fuse and the chip enter Power down or Power save sleep mode, the main clock source remains enabled. In these sleep modes, this will contribute significantly to the total current consumption. There are three alternative ways to avoid this:

• Disable OCDEN Fuse.

• Disable JTAGEN Fuse.

• Write one to the JTD bit in MCUCSR.

The TDO pin is left floating when the JTAG interface is enabled while the JTAG TAP controller is not shifting data.If the hardware connected to the TDO pin does not pull up the logic level, power consumption will increase. Notethat the TDI pin for the next device in the scan chain contains a pull-up that avoids this problem. Writing the JTD bitin the MCUCSR register to one or leaving the JTAG fuse unprogrammed disables the JTAG interface.



10.9.1 MCUCR – MCU Control RegisterThe MCU Control Register contains control bits for power management.

• Bit 7 – SE: Sleep EnableThe SE bit must be written to logic one to make the MCU enter the sleep mode when the SLEEP instruction is exe-cuted. To avoid the MCU entering the sleep mode unless it is the programmers purpose, it is recommended towrite the Sleep Enable (SE) bit to one just before the execution of the SLEEP instruction and to clear it immediatelyafter waking up.

• Bits 6:4 – SM2:0: Sleep Mode Select Bits 2, 1, and 0These bits select between the six available sleep modes as shown in Table 10-2.

Note: 1. Standby mode and Extended Standby mode are only available with external crystals or resonators.

Bit 7 6 5 4 3 2 1 0

SE SM2 SM1 SM0 ISC11 ISC10 ISC01 ISC00 MCUCR



Table 10-2. Sleep Mode Select

SM2 SM1 SM0 Sleep Mode

0 0 0 Idle

0 0 1 ADC Noise Reduction

0 1 0 Power-down

0 1 1 Power-save

1 0 0 Reserved

1 0 1 Reserved

1 1 0 Standby(1)

1 1 1 Extended Standby(1)


11. System Control and Reset

11.1 Resetting the AVRDuring Reset, all I/O Registers are set to their initial values, and the program starts execution from the Reset Vec-tor. The instruction placed at the Reset Vector must be a JMP – absolute jump – instruction to the reset handlingroutine. If the program never enables an interrupt source, the Interrupt Vectors are not used, and regular programcode can be placed at these locations. This is also the case if the Reset Vector is in the Application section whilethe Interrupt Vectors are in the Boot section or vice versa. The circuit diagram in Figure 10-1 shows the reset logic.“System and Reset Characteristics” on page 299 defines the electrical parameters of the reset circuitry.

The I/O ports of the AVR are immediately reset to their initial state when a reset source goes active. This does notrequire any clock source to be running.

After all reset sources have gone inactive, a delay counter is invoked, stretching the Internal Reset. This allows thepower to reach a stable level before normal operation starts. The time-out period of the delay counter is defined bythe user through the CKSEL Fuses. The different selections for the delay period are presented in “Clock Sources”on page 26.

11.2 Reset SourcesThe ATmega32A has five sources of reset:

• Power-on Reset. The MCU is reset when the supply voltage is below the Power-on Reset threshold (VPOT).

• External Reset. The MCU is reset when a low level is present on the RESET pin for longer than the minimum pulse length.

• Watchdog Reset. The MCU is reset when the Watchdog Timer period expires and the Watchdog is enabled.

• Brown-out Reset. The MCU is reset when the supply voltage VCC is below the Brown-out Reset threshold (VBOT) and the Brown-out Detector is enabled.

• JTAG AVR Reset. The MCU is reset as long as there is a logic one in the Reset Register, one of the scan chains of the JTAG system. Refer to the section “IEEE 1149.1 (JTAG) Boundary-scan” on page 233 for details.


Figure 11-1. Reset Logic

11.2.1 Power-on ResetA Power-on Reset (POR) pulse is generated by an On-chip detection circuit. The detection level is defined in “Sys-tem and Reset Characteristics” on page 299. The POR is activated whenever VCC is below the detection level. ThePOR circuit can be used to trigger the Start-up Reset, as well as to detect a failure in supply voltage.

A Power-on Reset (POR) circuit ensures that the device is reset from Power-on. Reaching the Power-on Resetthreshold voltage invokes the delay counter, which determines how long the device is kept in RESET after VCC rise.The RESET signal is activated again, without any delay, when VCC decreases below the detection level.

Figure 11-2. MCU Start-up, RESET Tied to VCC.

MCU Control and StatusRegister (MCUCSR)

BODENBODLEVEL

Delay Counters

CKSEL[3:0]

CKTIMEOUT

WD

RF

BO

RF

EX

TR

F

PO

RF

DATA BUS

ClockGenerator

SPIKEFILTER

Pull-up Resistor

JTR

F

JTAG ResetRegister

WatchdogOscillator

SUT[1:0]

WatchdogTimer

Reset Circuit

Brown-outReset Circuit

Power-onReset Circuit

INT

ER

NA

L R

ES

ET

CO

UN

TE

R R

ES

ET

V

RESET

TIME-OUT

INTERNALRESET

tTOUT

VPOT

VRST

CC


Figure 11-3. MCU Start-up, RESET Extended Externally

11.2.2 External ResetAn External Reset is generated by a low level on the RESET pin. Reset pulses longer than the minimum pulsewidth (see “System and Reset Characteristics” on page 299) will generate a reset, even if the clock is not running.Shorter pulses are not guaranteed to generate a reset. When the applied signal reaches the Reset Threshold Volt-age – VRST – on its positive edge, the delay counter starts the MCU after the Time-out period tTOUT has expired.

Figure 11-4. External Reset During Operation

11.2.3 Brown-out DetectionATmega32A has an On-chip Brown-out Detection (BOD) circuit for monitoring the VCC level during operation bycomparing it to a fixed trigger level. The trigger level for the BOD can be selected by the fuse BODLEVEL to be2.7V (BODLEVEL unprogrammed), or 4.0V (BODLEVEL programmed). The trigger level has a hysteresis toensure spike free Brown-out Detection. The hysteresis on the detection level should be interpreted as VBOT+ = VBOT

+ VHYST/2 and VBOT- = VBOT - VHYST/2.

The BOD circuit can be enabled/disabled by the fuse BODEN. When the BOD is enabled (BODEN programmed),and VCC decreases to a value below the trigger level (VBOT- in Figure 10-5), the Brown-out Reset is immediatelyactivated. When VCC increases above the trigger level (VBOT+ in Figure 10-5), the delay counter starts the MCUafter the Time-out period tTOUT has expired.

The BOD circuit will only detect a drop in VCC if the voltage stays below the trigger level for longer than tBOD givenin “System and Reset Characteristics” on page 299.

RESET

TIME-OUT

INTERNALRESET

tTOUT

VPOT

VRST

VCC

CC


Figure 11-5. Brown-out Reset During Operation

11.2.4 Watchdog ResetWhen the Watchdog times out, it will generate a short reset pulse of one CK cycle duration. On the falling edge ofthis pulse, the delay timer starts counting the Time-out period tTOUT. Refer to “Watchdog Timer” on page 42 fordetails.

Figure 11-6. Watchdog Reset During Operation

11.3 Internal Voltage ReferenceATmega32A features an internal bandgap reference. This reference is used for Brown-out Detection, and it can beused as an input to the Analog Comparator or the ADC. The 2.56V reference to the ADC is generated from theinternal bandgap reference.

11.3.1 Voltage Reference Enable Signals and Start-up TimeThe voltage reference has a start-up time that may influence the way it should be used. The start-up time is givenin “System and Reset Characteristics” on page 299. To save power, the reference is not always turned on. The ref-erence is on during the following situations:

1. When the BOD is enabled (by programming the BODEN Fuse).

2. When the bandgap reference is connected to the Analog Comparator (by setting the ACBG bit in ACSR).

3. When the ADC is enabled.

Thus, when the BOD is not enabled, after setting the ACBG bit or enabling the ADC, the user must always allowthe reference to start up before the output from the Analog Comparator or ADC is used. To reduce power con-sumption in Power-down mode, the user can avoid the three conditions above to ensure that the reference isturned off before entering Power-down mode.

VCC

RESET

TIME-OUT

INTERNALRESET

VBOT-VBOT+

tTOUT

CK

CC


11.4 Watchdog TimerThe Watchdog Timer is clocked from a separate On-chip Oscillator which runs at 1MHz. This is the typical value atVCC = 5V. See characterization data for typical values at other VCC levels. By controlling the Watchdog Timer pres-caler, the Watchdog Reset interval can be adjusted as shown in Table 10-1 on page 44. The WDR – WatchdogReset – instruction resets the Watchdog Timer. The Watchdog Timer is also reset when it is disabled and when aChip Reset occurs. Eight different clock cycle periods can be selected to determine the reset period. If the resetperiod expires without another Watchdog Reset, the ATmega32A resets and executes from the Reset Vector. Fortiming details on the Watchdog Reset, refer to page 41.

To prevent unintentional disabling of the Watchdog, a special turn-off sequence must be followed when the Watch-dog is disabled. Refer to the description of the Watchdog Timer Control Register for details.

Figure 11-7. Watchdog Timer

WATCHDOGOSCILLATOR



11.5.1 MCUCSR – MCU Control and Status RegisterThe MCU Control and Status Register provides information on which reset source caused an MCU Reset.

• Bit 4 – JTRF: JTAG Reset FlagThis bit is set if a reset is being caused by a logic one in the JTAG Reset Register selected by the JTAG instructionAVR_RESET. This bit is reset by a Power-on Reset, or by writing a logic zero to the flag.

• Bit 3 – WDRF: Watchdog Reset FlagThis bit is set if a Watchdog Reset occurs. The bit is reset by a Power-on Reset, or by writing a logic zero to theflag.

• Bit 2 – BORF: Brown-out Reset FlagThis bit is set if a Brown-out Reset occurs. The bit is reset by a Power-on Reset, or by writing a logic zero to theflag.

• Bit 1 – EXTRF: External Reset FlagThis bit is set if an External Reset occurs. The bit is reset by a Power-on Reset, or by writing a logic zero to the flag.

• Bit 0 – PORF: Power-on Reset FlagThis bit is set if a Power-on Reset occurs. The bit is reset only by writing a logic zero to the flag.

To make use of the Reset Flags to identify a reset condition, the user should read and then reset the MCUCSR asearly as possible in the program. If the register is cleared before another reset occurs, the source of the reset canbe found by examining the Reset Flags.

11.5.2 WDTCR – Watchdog Timer Control Register

• Bits 7:5 – Reserved BitsThese bits are reserved bits in the ATmega32A and will always read as zero.

• Bit 4 – WDTOE: Watchdog Turn-off EnableThis bit must be set when the WDE bit is written to logic zero. Otherwise, the Watchdog will not be disabled. Oncewritten to one, hardware will clear this bit after four clock cycles. Refer to the description of the WDE bit for aWatchdog disable procedure.

• Bit 3 – WDE: Watchdog EnableWhen the WDE is written to logic one, the Watchdog Timer is enabled, and if the WDE is written to logic zero, theWatchdog Timer function is disabled. WDE can only be cleared if the WDTOE bit has logic level one. To disable anenabled Watchdog Timer, the following procedure must be followed:

Bit 7 6 5 4 3 2 1 0

JTD ISC2 – JTRF WDRF BORF EXTRF PORF MCUCSR

Read/Write R/W R/W R R/W R/W R/W R/W R/W

Initial Value 0 0 0 See Bit Description

Bit 7 6 5 4 3 2 1 0

– – – WDTOE WDE WDP2 WDP1 WDP0 WDTCR

Read/Write R R R R/W R/W R/W R/W R/W



1. In the same operation, write a logic one to WDTOE and WDE. A logic one must be written to WDE even though it is set to one before the disable operation starts.

2. Within the next four clock cycles, write a logic 0 to WDE. This disables the Watchdog.

• Bits 2:0 – WDP2, WDP1, WDP0: Watchdog Timer Prescaler 2, 1, and 0The WDP2, WDP1, and WDP0 bits determine the Watchdog Timer prescaling when the Watchdog Timer isenabled. The different prescaling values and their corresponding Timeout Periods are shown in Table 10-1.

The following code example shows one assembly and one C function for turning off the WDT. The exampleassumes that interrupts are controlled (for example by disabling interrupts globally) so that no interrupts will occurduring execution of these functions.

Table 11-1. Watchdog Timer Prescale Select

WDP2 WDP1 WDP0Number of WDT Oscillator Cycles

Typical Time-out at VCC = 3.0V

Typical Time-out at VCC = 5.0V

0 0 0 16K (16,384) 17.1ms 16.3ms

0 0 1 32K (32,768) 34.3ms 32.5ms

0 1 0 64K (65,536) 68.5ms 65ms

0 1 1 128K (131,072) 0.14s 0.13s

1 0 0 256K (262,144) 0.27s 0.26s

1 0 1 512K (524,288) 0.55s 0.52s

1 1 0 1,024K (1,048,576) 1.1s 1.0s

1 1 1 2,048K (2,097,152) 2.2s 2.1s


WDT_off:

; reset WDT

wdr

; Write logical one to WDTOE and WDE

in r16, WDTCR

ori r16, (1<<WDTOE)|(1<<WDE)

out WDTCR, r16

; Turn off WDT

ldi r16, (0<<WDE)

out WDTCR, r16

ret

C Code Example

void WDT_off(void)

{

/* reset WDT */

_WDR();

/* Write logical one to WDTOE and WDE */

WDTCR |= (1<<WDTOE) | (1<<WDE);

/* Turn off WDT */

WDTCR = 0x00;

}


12. InterruptsThis section describes the specifics of the interrupt handling as performed in ATmega32A. For a general explana-tion of the AVR interrupt handling, refer to “Reset and Interrupt Handling” on page 13.

12.1 Interrupt Vectors in ATmega32A

Notes: 1. When the BOOTRST fuse is programmed, the device will jump to the Boot Loader address at reset, see “Boot Loader Support – Read-While-Write Self-Programming” on page 252.

2. When the IVSEL bit in GICR is set, interrupt vectors will be moved to the start of the Boot Flash section. The address of each Interrupt Vector will then be the address in this table added to the start address of the Boot Flash section.

Table 11-2 shows Reset and Interrupt Vectors placement for the various combinations of BOOTRST and IVSELsettings. If the program never enables an interrupt source, the Interrupt Vectors are not used, and regular programcode can be placed at these locations. This is also the case if the Reset Vector is in the Application section whilethe Interrupt Vectors are in the Boot section or vice versa.

Table 12-1. Reset and Interrupt Vectors

Vector No.Program

Address(2) Source Interrupt Definition

1 $000(1) RESET External Pin, Power-on Reset, Brown-out Reset, Watchdog Reset, and JTAG AVR Reset

2 $002 INT0 External Interrupt Request 0



5 $008 TIMER2 COMP Timer/Counter2 Compare Match

6 $00A TIMER2 OVF Timer/Counter2 Overflow

7 $00C TIMER1 CAPT Timer/Counter1 Capture Event

8 $00E TIMER1 COMPA Timer/Counter1 Compare Match A

9 $010 TIMER1 COMPB Timer/Counter1 Compare Match B

10 $012 TIMER1 OVF Timer/Counter1 Overflow

11 $014 TIMER0 COMP Timer/Counter0 Compare Match

12 $016 TIMER0 OVF Timer/Counter0 Overflow

13 $018 SPI, STC Serial Transfer Complete

14 $01A USART, RXC USART, Rx Complete

15 $01C USART, UDRE USART Data Register Empty

16 $01E USART, TXC USART, Tx Complete

17 $020 ADC ADC Conversion Complete

18 $022 EE_RDY EEPROM Ready

19 $024 ANA_COMP Analog Comparator

20 $026 TWI Two-wire Serial Interface

21 $028 SPM_RDY Store Program Memory Ready


Note: 1. The Boot Reset Address is shown in Table 25-6 on page 263. For the BOOTRST Fuse “1” means unprogrammed while “0” means programmed.

The most typical and general program setup for the Reset and Interrupt Vector Addresses in ATmega32A is:

Address Labels Code Comments

$000 jmp RESET ; Reset Handler

$002 jmp EXT_INT0 ; IRQ0 Handler



$008 jmp TIM2_COMP ; Timer2 Compare Handler

$00A jmp TIM2_OVF ; Timer2 Overflow Handler

$00C jmp TIM1_CAPT ; Timer1 Capture Handler

$00E jmp TIM1_COMPA ; Timer1 CompareA Handler

$010 jmp TIM1_COMPB ; Timer1 CompareB Handler

$012 jmp TIM1_OVF ; Timer1 Overflow Handler

$014 jmp TIM0_COMP ; Timer0 Compare Handler

$016 jmp TIM0_OVF ; Timer0 Overflow Handler

$018 jmp SPI_STC ; SPI Transfer Complete Handler

$01A jmp USART_RXC ; USART RX Complete Handler

$01C jmp USART_UDRE ; UDR Empty Handler

$01E jmp USART_TXC ; USART TX Complete Handler

$020 jmp ADC ; ADC Conversion Complete Handler

$022 jmp EE_RDY ; EEPROM Ready Handler

$024 jmp ANA_COMP ; Analog Comparator Handler

$026 jmp TWI ; Two-wire Serial Interface Handler

$028 jmp SPM_RDY ; Store Program Memory Ready Handler

;

$02A RESET: ldi r16,high(RAMEND); Main program start

$02B out SPH,r16 ; Set Stack Pointer to top of RAM

$02C ldi r16,low(RAMEND)

$02D out SPL,r16

$02E sei ; Enable interrupts

$02F <instr> xxx

:. :. :.

Table 12-2. Reset and Interrupt Vectors Placement(1)

BOOTRST IVSEL Reset address Interrupt Vectors Start Address

1 0 $0000 $0002

1 1 $0000 Boot Reset Address + $0002

0 0 Boot Reset Address $0002

0 1 Boot Reset Address Boot Reset Address + $0002


When the BOOTRST Fuse is unprogrammed, the Boot section size set to 4Kbytes and the IVSEL bit in the GICRRegister is set before any interrupts are enabled, the most typical and general program setup for the Reset andInterrupt Vector Addresses is:


$000 RESET: ldi r16,high(RAMEND); Main program start

$001 out SPH,r16 ; Set Stack Pointer to top of RAM

$002 ldi r16,low(RAMEND)

$003 out SPL,r16

$004 sei ; Enable interrupts

$005 <instr> xxx

;

.org $3802



:. :.. : ;


When the BOOTRST Fuse is programmed and the Boot section size set to 4Kbytes, the most typical and generalprogram setup for the Reset and Interrupt Vector Addresses is:


.org $002



:. :.. : ;


;

.org $3800$3800 RESET: ldi r16,high(RAMEND); Main program start

$3801 out SPH,r16 ; Set Stack Pointer to top of RAM

$3802 ldi r16,low(RAMEND)

$3803 out SPL,r16

$3804 sei ; Enable interrupts

$3805 <instr> xxx

When the BOOTRST Fuse is programmed, the Boot section size set to 4Kbytes and the IVSEL bit in the GICRRegister is set before any interrupts are enabled, the most typical and general program setup for the Reset andInterrupt Vector Addresses is:


.org $3800$3800 jmp RESET ; Reset handler$3802 jmp EXT_INT0 ; IRQ0 Handler


:. :.. : ;


;

$382A RESET: ldi r16,high(RAMEND); Main program start

$382B out SPH,r16 ; Set Stack Pointer to top of RAM

$382C ldi r16,low(RAMEND)

$382D out SPL,r16


$382E sei ; Enable interrupts

$382F <instr> xxx

12.1.1 Moving Interrupts Between Application and Boot SpaceThe General Interrupt Control Register controls the placement of the Interrupt Vector table.


12.2.1 GICR – General Interrupt Control Register

• Bit 1 – IVSEL: Interrupt Vector SelectWhen the IVSEL bit is cleared (zero), the Interrupt Vectors are placed at the start of the Flash memory. When thisbit is set (one), the interrupt vectors are moved to the beginning of the Boot Loader section of the Flash. The actualaddress of the start of the Boot Flash section is determined by the BOOTSZ fuses. Refer to the section “BootLoader Support – Read-While-Write Self-Programming” on page 252 for details. To avoid unintentional changes ofInterrupt Vector tables, a special write procedure must be followed to change the IVSEL bit:

1. Write the Interrupt Vector Change Enable (IVCE) bit to one.

2. Within four cycles, write the desired value to IVSEL while writing a zero to IVCE.

Interrupts will automatically be disabled while this sequence is executed. Interrupts are disabled in the cycle IVCEis set, and they remain disabled until after the instruction following the write to IVSEL. If IVSEL is not written, inter-rupts remain disabled for four cycles. The I-bit in the Status Register is unaffected by the automatic disabling.

Note: If Interrupt Vectors are placed in the Boot Loader section and Boot Lock bit BLB02 is programmed, interrupts are dis-abled while executing from the Application section. If Interrupt Vectors are placed in the Application section and Boot Lock bit BLB12 is programed, interrupts are disabled while executing from the Boot Loader section. Refer to the sec-tion “Boot Loader Support – Read-While-Write Self-Programming” on page 252 for details on Boot Lock bits.

• Bit 0 – IVCE: Interrupt Vector Change EnableThe IVCE bit must be written to logic one to enable change of the IVSEL bit. IVCE is cleared by hardware fourcycles after it is written or when IVSEL is written. Setting the IVCE bit will disable interrupts, as explained in theIVSEL description above. See Code Example below.

Bit 7 6 5 4 3 2 1 0

INT1 INT0 INT2 – – – IVSEL IVCE GICR

Read/Write R/W R/W R/W R R R R/W R/W




Move_interrupts:

; Enable change of interrupt vectors

ldi r16, (1<<IVCE)

out GICR, r16

; Move interrupts to boot Flash section

ldi r16, (1<<IVSEL)

out GICR, r16

ret

C Code Example

void Move_interrupts(void)

{

/* Enable change of interrupt vectors */

GICR = (1<<IVCE);

/* Move interrupts to boot Flash section */

GICR = (1<<IVSEL);

}


13. I/O Ports

13.1 OverviewAll AVR ports have true Read-Modify-Write functionality when used as general digital I/O ports. This means thatthe direction of one port pin can be changed without unintentionally changing the direction of any other pin with theSBI and CBI instructions. The same applies when changing drive value (if configured as output) or enabling/dis-abling of pull-up resistors (if configured as input). Each output buffer has symmetrical drive characteristics withboth high sink and source capability. The pin driver is strong enough to drive LED displays directly. All port pinshave individually selectable pull-up resistors with a supply-voltage invariant resistance. All I/O pins have protectiondiodes to both VCC and Ground as indicated in Figure 12-1. Refer to “Electrical Characteristics” on page 296 for acomplete list of parameters.

Figure 13-1. I/O Pin Equivalent Schematic

All registers and bit references in this section are written in general form. A lower case “x” represents the number-ing letter for the port, and a lower case “n” represents the bit number. However, when using the register or bitdefines in a program, the precise form must be used, that is, PORTB3 for bit no. 3 in Port B, here documented gen-erally as PORTxn. The physical I/O Registers and bit locations are listed in “Register Description” on page 66.

Three I/O memory address locations are allocated for each port, one each for the Data Register – PORTx, DataDirection Register – DDRx, and the Port Input Pins – PINx. The Port Input Pins I/O location is read only, while theData Register and the Data Direction Register are read/write. In addition, the Pull-up Disable – PUD bit in SFIORdisables the pull-up function for all pins in all ports when set.

Using the I/O port as General Digital I/O is described in “Ports as General Digital I/O” on page 51. Most port pinsare multiplexed with alternate functions for the peripheral features on the device. How each alternate function inter-feres with the port pin is described in “Alternate Port Functions” on page 55. Refer to the individual module sectionsfor a full description of the alternate functions.

Note that enabling the alternate function of some of the port pins does not affect the use of the other pins in the portas general digital I/O.

Cpin

Logic

Rpu

See Figure 23"General Digital I/O" for

Details

Pxn


13.2 Ports as General Digital I/OThe ports are bi-directional I/O ports with optional internal pull-ups. Figure 12-2 shows a functional description ofone I/O-port pin, here generically called Pxn.

Figure 13-2. General Digital I/O(1)

Note: 1. WPx, WDx, RRx, RPx, and RDx are common to all pins within the same port. clkI/O, SLEEP, and PUD are commonto all ports.

13.2.1 Configuring the PinEach port pin consists of three register bits: DDxn, PORTxn, and PINxn. As shown in “Register Description” onpage 66, the DDxn bits are accessed at the DDRx I/O address, the PORTxn bits at the PORTx I/O address, andthe PINxn bits at the PINx I/O address.

The DDxn bit in the DDRx Register selects the direction of this pin. If DDxn is written logic one, Pxn is configuredas an output pin. If DDxn is written logic zero, Pxn is configured as an input pin.

If PORTxn is written logic one when the pin is configured as an input pin, the pull-up resistor is activated. To switchthe pull-up resistor off, PORTxn has to be written logic zero or the pin has to be configured as an output pin. Theport pins are tri-stated when a reset condition becomes active, even if no clocks are running.

If PORTxn is written logic one when the pin is configured as an output pin, the port pin is driven high (one). IfPORTxn is written logic zero when the pin is configured as an output pin, the port pin is driven low (zero).

When switching between tri-state ({DDxn, PORTxn} = 0b00) and output high ({DDxn, PORTxn} = 0b11), an inter-mediate state with either pull-up enabled ({DDxn, PORTxn} = 0b01) or output low ({DDxn, PORTxn} = 0b10) mustoccur. Normally, the pull-up enabled state is fully acceptable, as a high-impedant environment will not notice thedifference between a strong high driver and a pull-up. If this is not the case, the PUD bit in the SFIOR Register canbe set to disable all pull-ups in all ports.

clk

RPx

RRx

WPx

RDx

WDx

PUD

SYNCHRONIZER

WDx: WRITE DDRx

WPx: WRITE PORTxRRx: READ PORTx REGISTERRPx: READ PORTx PIN

PUD: PULLUP DISABLE

clkI/O: I/O CLOCK

RDx: READ DDRx

D

L

Q

Q

RESET

RESET

Q

QD

Q

Q D

CLR

PORTxn

Q

Q D

CLR

DDxn

PINxn

DAT

A B

US

SLEEP

SLEEP: SLEEP CONTROL

Pxn

I/O


Switching between input with pull-up and output low generates the same problem. The user must use either the tri-state ({DDxn, PORTxn} = 0b00) or the output high state ({DDxn, PORTxn} = 0b11) as an intermediate step.

Table 12-1 summarizes the control signals for the pin value.

13.2.2 Reading the Pin ValueIndependent of the setting of Data Direction bit DDxn, the port pin can be read through the PINxn Register bit. Asshown in Figure 12-2, the PINxn Register bit and the preceding latch constitute a synchronizer. This is needed toavoid metastability if the physical pin changes value near the edge of the internal clock, but it also introduces adelay. Figure 12-3 shows a timing diagram of the synchronization when reading an externally applied pin value.The maximum and minimum propagation delays are denoted tpd,max and tpd,min respectively.

Figure 13-3. Synchronization when Reading an Externally Applied Pin Value

Consider the clock period starting shortly after the first falling edge of the system clock. The latch is closed whenthe clock is low, and goes transparent when the clock is high, as indicated by the shaded region of the “SYNCLATCH” signal. The signal value is latched when the system clock goes low. It is clocked into the PINxn Register atthe succeeding positive clock edge. As indicated by the two arrows tpd,max and tpd,min, a single signal transition onthe pin will be delayed between ½ and 1½ system clock period depending upon the time of assertion.

When reading back a software assigned pin value, a nop instruction must be inserted as indicated in Figure 12-4.The out instruction sets the “SYNC LATCH” signal at the positive edge of the clock. In this case, the delay tpd

through the synchronizer is one system clock period.

Table 13-1. Port Pin Configurations

DDxn PORTxnPUD

(in SFIOR) I/O Pull-up Comment

0 0 X Input No Tri-state (Hi-Z)

0 1 0 Input Yes Pxn will source current if ext. pulled low.

0 1 1 Input No Tri-state (Hi-Z)

1 0 X Output No Output Low (Sink)

1 1 X Output No Output High (Source)

SYSTEM CLK

INSTRUCTIONS

SYNC LATCH

PINxn

r17

in r17, PINx

0xFF0x00

tpd, max

XXXXXX

tpd, min


Figure 13-4. Synchronization when Reading a Software Assigned Pin Value

The following code example shows how to set port B pins 0 and 1 high, 2 and 3 low, and define the port pins from4 to 7 as input with pull-ups assigned to port pins 6 and 7. The resulting pin values are read back again, but as pre-viously discussed, a nop instruction is included to be able to read back the value recently assigned to some of thepins.

nop in r17, PINx

0xFF

0x00 0xFF

tpd

out PORTx, r16

SYSTEM CLK

r16

INSTRUCTIONS

SYNC LATCH

PINxn

r17


Note: 1. For the assembly program, two temporary registers are used to minimize the time from pull-ups are set on pins 0, 1, 6, and 7, until the direction bits are correctly set, defining bit 2 and 3 as low and redefining bits 0 and 1 as strong high drivers.

13.2.3 Digital Input Enable and Sleep ModesAs shown in Figure 12-2, the digital input signal can be clamped to ground at the input of the schmitt-trigger. Thesignal denoted SLEEP in the figure, is set by the MCU Sleep Controller in Power-down mode, Power-save mode,Standby mode, and Extended Standby mode to avoid high power consumption if some input signals are left float-ing, or have an analog signal level close to VCC/2.

SLEEP is overridden for port pins enabled as External Interrupt pins. If the External Interrupt Request is notenabled, SLEEP is active also for these pins. SLEEP is also overridden by various other alternate functions asdescribed in “Alternate Port Functions” on page 55.

If a logic high level (“one”) is present on an Asynchronous External Interrupt pin configured as “Interrupt on RisingEdge, Falling Edge, or Any Logic Change on Pin” while the External Interrupt is not enabled, the correspondingExternal Interrupt Flag will be set when resuming from the above mentioned sleep modes, as the clamping in thesesleep modes produces the requested logic change.

13.2.4 Unconnected pinsIf some pins are unused, it is recommended to ensure that these pins have a defined level. Even though most ofthe digital inputs are disabled in the deep sleep modes as described above, floating inputs should be avoided to

Assembly Code Example(1)

:.

; Define pull-ups and set outputs high

; Define directions for port pins

ldi r16,(1<<PB7)|(1<<PB6)|(1<<PB1)|(1<<PB0)

ldi r17,(1<<DDB3)|(1<<DDB2)|(1<<DDB1)|(1<<DDB0)

out PORTB,r16

out DDRB,r17

; Insert nop for synchronization

nop

; Read port pins

in r16,PINB

:.

C Code Example(1)

unsigned char i;

:.

/* Define pull-ups and set outputs high */

/* Define directions for port pins */

PORTB = (1<<PB7)|(1<<PB6)|(1<<PB1)|(1<<PB0);

DDRB = (1<<DDB3)|(1<<DDB2)|(1<<DDB1)|(1<<DDB0);

/* Insert nop for synchronization*/

_NOP();

/* Read port pins */

i = PINB;

:.


reduce current consumption in all other modes where the digital inputs are enabled (Reset, Active mode and Idlemode).

The simplest method to ensure a defined level of an unused pin, is to enable the internal pullup. In this case, thepullup will be disabled during reset. If low power consumption during reset is important, it is recommended to usean external pullup or pulldown. Connecting unused pins directly to VCC or GND is not recommended, since thismay cause excessive currents if the pin is accidentally configured as an output.

13.3 Alternate Port FunctionsMost port pins have alternate functions in addition to being General Digital I/Os. Figure 12-5 shows how the portpin control signals from the simplified Figure 12-2 can be overridden by alternate functions. The overriding signalsmay not be present in all port pins, but the figure serves as a generic description applicable to all port pins in theAVR microcontroller family.

Figure 13-5. Alternate Port Functions(1)

Note: 1. WPx, WDx, RRx, RPx, and RDx are common to all pins within the same port. clkI/O, SLEEP, and PUD are commonto all ports. All other signals are unique for each pin.

Table 12-2 summarizes the function of the overriding signals. The pin and port indexes from Figure 12-5 are notshown in the succeeding tables. The overriding signals are generated internally in the modules having the alternatefunction.

clk

RPx

RRx

WPx

RDx

WDx

PUD

SYNCHRONIZER

WDx: WRITE DDRx

WPx: WRITE PORTxRRx: READ PORTx REGISTER

RPx: READ PORTx PIN

PUD: PULLUP DISABLE

clkI/O: I/O CLOCK

RDx: READ DDRx

D

L

Q

Q

SET

CLR

0

1

0

1

0

1

DIxn

AIOxn

DIEOExn

PVOVxn

PVOExn

DDOVxn

DDOExn

PUOExn

PUOVxn

PUOExn: Pxn PULL-UP OVERRIDE ENABLEPUOVxn: Pxn PULL-UP OVERRIDE VALUEDDOExn: Pxn DATA DIRECTION OVERRIDE ENABLEDDOVxn: Pxn DATA DIRECTION OVERRIDE VALUEPVOExn: Pxn PORT VALUE OVERRIDE ENABLEPVOVxn: Pxn PORT VALUE OVERRIDE VALUE

DIxn: DIGITAL INPUT PIN n ON PORTxAIOxn: ANALOG INPUT/OUTPUT PIN n ON PORTx

RESET

RESET

Q

Q D

CLR

Q

Q D

CLR

Q

QD

CLR

PINxn

PORTxn

DDxn

DAT

A B

US

0

1DIEOVxn

SLEEP

DIEOExn: Pxn DIGITAL INPUT-ENABLE OVERRIDE ENABLEDIEOVxn: Pxn DIGITAL INPUT-ENABLE OVERRIDE VALUESLEEP: SLEEP CONTROL

Pxn

I/O


The following subsections shortly describe the alternate functions for each port, and relate the overriding signals tothe alternate function. Refer to the alternate function description for further details.

13.3.1 Alternate Functions of Port APort A has an alternate function as analog input for the ADC as shown in Table 12-3. If some Port A pins are con-figured as outputs, it is essential that these do not switch when a conversion is in progress. This might corrupt theresult of the conversion.

Table 13-2. Generic Description of Overriding Signals for Alternate Functions

Signal Name Full Name Description

PUOE Pull-up Override Enable

If this signal is set, the pull-up enable is controlled by the PUOV signal. If this signal is cleared, the pull-up is enabled when {DDxn, PORTxn, PUD} = 0b010.

PUOV Pull-up Override Value If PUOE is set, the pull-up is enabled/disabled when PUOV is set/cleared, regardless of the setting of the DDxn, PORTxn, and PUD Register bits.

DDOE Data Direction Override Enable

If this signal is set, the Output Driver Enable is controlled by the DDOV signal. If this signal is cleared, the Output driver is enabled by the DDxn Register bit.

DDOV Data Direction Override Value

If DDOE is set, the Output Driver is enabled/disabled when DDOV is set/cleared, regardless of the setting of the DDxn Register bit.

PVOE Port Value Override Enable

If this signal is set and the Output Driver is enabled, the port value is controlled by the PVOV signal. If PVOE is cleared, and the Output Driver is enabled, the port Value is controlled by the PORTxn Register bit.

PVOV Port Value Override Value

If PVOE is set, the port value is set to PVOV, regardless of the setting of the PORTxn Register bit.

DIEOE Digital Input Enable Override Enable

If this bit is set, the Digital Input Enable is controlled by the DIEOV signal. If this signal is cleared, the Digital Input Enable is determined by MCU-state (Normal Mode, sleep modes).

DIEOV Digital Input Enable Override Value

If DIEOE is set, the Digital Input is enabled/disabled when DIEOV is set/cleared, regardless of the MCU state (Normal Mode, sleep modes).

DI Digital Input This is the Digital Input to alternate functions. In the figure, the signal is connected to the output of the schmitt trigger but before the synchronizer. Unless the Digital Input is used as a clock source, the module with the alternate function will use its own synchronizer.

AIO Analog Input/ output This is the Analog Input/output to/from alternate functions. The signal is connected directly to the pad, and can be used bi-directionally.


Table 12-4 and Table 12-5 relate the alternate functions of Port A to the overriding signals shown in Figure 12-5 onpage 56.

Table 13-3. Port A Pins Alternate Functions

Port Pin Alternate Function

PA7 ADC7 (ADC input channel 7)








Table 13-4. Overriding Signals for Alternate Functions in PA7:PA4

Signal Name PA7/ADC7 PA6/ADC6 PA5/ADC5 PA4/ADC4

PUOE 0 0 0 0

PUOV 0 0 0 0

DDOE 0 0 0 0

DDOV 0 0 0 0

PVOE 0 0 0 0

PVOV 0 0 0 0

DIEOE 0 0 0 0

DIEOV 0 0 0 0

DI – – – –

AIO ADC7 INPUT ADC6 INPUT ADC5 INPUT ADC4 INPUT

Table 13-5. Overriding Signals for Alternate Functions in PA3:PA0

Signal Name PA3/ADC3 PA2/ADC2 PA1/ADC1 PA0/ADC0

PUOE 0 0 0 0

PUOV 0 0 0 0

DDOE 0 0 0 0

DDOV 0 0 0 0

PVOE 0 0 0 0

PVOV 0 0 0 0

DIEOE 0 0 0 0

DIEOV 0 0 0 0

DI – – – –

AIO ADC3 INPUT ADC2 INPUT ADC1 INPUT ADC0 INPUT


13.3.2 Alternate Functions of Port BThe Port B pins with alternate functions are shown in Table 12-6.

The alternate pin configuration is as follows:

• SCK – Port B, Bit 7SCK: Master Clock output, Slave Clock input pin for SPI. When the SPI is enabled as a Slave, this pin is configuredas an input regardless of the setting of DDB7. When the SPI is enabled as a Master, the data direction of this pin iscontrolled by DDB7. When the pin is forced by the SPI to be an input, the pull-up can still be controlled by thePORTB7 bit.

• MISO – Port B, Bit 6MISO: Master Data input, Slave Data output pin for SPI. When the SPI is enabled as a Master, this pin is config-ured as an input regardless of the setting of DDB6. When the SPI is enabled as a Slave, the data direction of thispin is controlled by DDB6. When the pin is forced by the SPI to be an input, the pull-up can still be controlled by thePORTB6 bit.

• MOSI – Port B, Bit 5MOSI: SPI Master Data output, Slave Data input for SPI. When the SPI is enabled as a Slave, this pin is configuredas an input regardless of the setting of DDB5. When the SPI is enabled as a Master, the data direction of this pin iscontrolled by DDB5. When the pin is forced by the SPI to be an input, the pull-up can still be controlled by thePORTB5 bit.

• SS – Port B, Bit 4SS: Slave Select input. When the SPI is enabled as a Slave, this pin is configured as an input regardless of the set-ting of DDB4. As a Slave, the SPI is activated when this pin is driven low. When the SPI is enabled as a Master, thedata direction of this pin is controlled by DDB4. When the pin is forced by the SPI to be an input, the pull-up can stillbe controlled by the PORTB4 bit.

• AIN1/OC0 – Port B, Bit 3AIN1, Analog Comparator Negative Input. Configure the port pin as input with the internal pull-up switched off toavoid the digital port function from interfering with the function of the analog comparator.

Table 13-6. Port B Pins Alternate Functions

Port Pin Alternate Functions

PB7 SCK (SPI Bus Serial Clock)

PB6 MISO (SPI Bus Master Input/Slave Output)

PB5 MOSI (SPI Bus Master Output/Slave Input)

PB4 SS (SPI Slave Select Input)

PB3AIN1 (Analog Comparator Negative Input)OC0 (Timer/Counter0 Output Compare Match Output)

PB2AIN0 (Analog Comparator Positive Input)INT2 (External Interrupt 2 Input)

PB1 T1 (Timer/Counter1 External Counter Input)

PB0T0 (Timer/Counter0 External Counter Input)

XCK (USART External Clock Input/Output)


OC0, Output Compare Match output: The PB3 pin can serve as an external output for the Timer/Counter0 Com-pare Match. The PB3 pin has to be configured as an output (DDB3 set (one)) to serve this function. The OC0 pin isalso the output pin for the PWM mode timer function.

• AIN0/INT2 – Port B, Bit 2AIN0, Analog Comparator Positive input. Configure the port pin as input with the internal pull-up switched off toavoid the digital port function from interfering with the function of the Analog Comparator.

INT2, External Interrupt Source 2: The PB2 pin can serve as an external interrupt source to the MCU.

• T1 – Port B, Bit 1T1, Timer/Counter1 Counter Source.

• T0/XCK – Port B, Bit 0T0, Timer/Counter0 Counter Source.

XCK, USART External Clock. The Data Direction Register (DDB0) controls whether the clock is output (DDB0 set)or input (DDB0 cleared). The XCK pin is active only when the USART operates in Synchronous mode.

Table 12-7 and Table 12-8 relate the alternate functions of Port B to the overriding signals shown in Figure 12-5 onpage 56. SPI MSTR INPUT and SPI SLAVE OUTPUT constitute the MISO signal, while MOSI is divided into SPIMSTR OUTPUT and SPI SLAVE INPUT.

Table 13-7. Overriding Signals for Alternate Functions in PB7:PB4

SignalName PB7/SCK PB6/MISO PB5/MOSI PB4/SS

PUOE SPE • MSTR SPE • MSTR SPE • MSTR SPE • MSTR

PUOV PORTB7 • PUD PORTB6 • PUD PORTB5 • PUD PORTB4 • PUD

DDOE SPE • MSTR SPE • MSTR SPE • MSTR SPE • MSTR

DDOV 0 0 0 0

PVOE SPE • MSTR SPE • MSTR SPE • MSTR 0

PVOV SCK OUTPUT SPI SLAVE OUTPUT SPI MSTR OUTPUT 0

DIEOE 0 0 0 0

DIEOV 0 0 0 0

DI SCK INPUT SPI MSTR INPUT SPI SLAVE INPUT SPI SS

AIO – – – –


13.3.3 Alternate Functions of Port CThe Port C pins with alternate functions are shown in Table 12-9. If the JTAG interface is enabled, the pull-up resis-tors on pins PC5(TDI), PC3(TMS) and PC2(TCK) will be activated even if a reset occurs.


• TOSC2 – Port C, Bit 7TOSC2, Timer Oscillator pin 2: When the AS2 bit in ASSR is set (one) to enable asynchronous clocking ofTimer/Counter2, pin PC7 is disconnected from the port, and becomes the inverting output of the Oscillator ampli-fier. In this mode, a Crystal Oscillator is connected to this pin, and the pin can not be used as an I/O pin.

• TOSC1 – Port C, Bit 6TOSC1, Timer Oscillator pin 1: When the AS2 bit in ASSR is set (one) to enable asynchronous clocking ofTimer/Counter2, pin PC6 is disconnected from the port, and becomes the input of the inverting Oscillator amplifier.In this mode, a Crystal Oscillator is connected to this pin, and the pin can not be used as an I/O pin.

• TDI – Port C, Bit 5TDI, JTAG Test Data In: Serial input data to be shifted in to the Instruction Register or Data Register (scan chains).When the JTAG interface is enabled, this pin can not be used as an I/O pin.

Table 13-8. Overriding Signals for Alternate Functions in PB3:PB0

Signal Name PB3/OC0/AIN1 PB2/INT2/AIN0 PB1/T1 PB0/T0/XCK

PUOE 0 0 0 0

PUOV 0 0 0 0

DDOE 0 0 0 0

DDOV 0 0 0 0

PVOE OC0 ENABLE 0 0 UMSEL

PVOV OC0 0 0 XCK OUTPUT

DIEOE 0 INT2 ENABLE 0 0

DIEOV 0 1 0 0

DI – INT2 INPUT T1 INPUT XCK INPUT/T0 INPUT

AIO AIN1 INPUT AIN0 INPUT – –

Table 13-9. Port C Pins Alternate Functions


PC7 TOSC2 (Timer Oscillator Pin 2)

PC6 TOSC1 (Timer Oscillator Pin 1)

PC5 TDI (JTAG Test Data In)

PC4 TDO (JTAG Test Data Out)

PC3 TMS (JTAG Test Mode Select)

PC2 TCK (JTAG Test Clock)

PC1 SDA (Two-wire Serial Bus Data Input/Output Line)

PC0 SCL (Two-wire Serial Bus Clock Line)


• TDO – Port C, Bit 4TDO, JTAG Test Data Out: Serial output data from Instruction Register or Data Register. When the JTAG interfaceis enabled, this pin can not be used as an I/O pin.

The TD0 pin is tri-stated unless TAP states that shifts out data are entered.

• TMS – Port C, Bit 3TMS, JTAG Test Mode Select: This pin is used for navigating through the TAP-controller state machine. When theJTAG interface is enabled, this pin can not be used as an I/O pin.

• TCK – Port C, Bit 2TCK, JTAG Test Clock: JTAG operation is synchronous to TCK. When the JTAG interface is enabled, this pin cannot be used as an I/O pin.

• SDA – Port C, Bit 1SDA, Two-wire Serial Interface Data: When the TWEN bit in TWCR is set (one) to enable the Two-wire Serial Inter-face, pin PC1 is disconnected from the port and becomes the Serial Data I/O pin for the Two-wire Serial Interface.In this mode, there is a spike filter on the pin to suppress spikes shorter than 50 ns on the input signal, and the pinis driven by an open drain driver with slew-rate limitation. When this pin is used by the Two-wire Serial Interface,the pull-up can still be controlled by the PORTC1 bit.

• SCL – Port C, Bit 0SCL, Two-wire Serial Interface Clock: When the TWEN bit in TWCR is set (one) to enable the Two-wire SerialInterface, pin PC0 is disconnected from the port and becomes the Serial Clock I/O pin for the Two-wire Serial Inter-face. In this mode, there is a spike filter on the pin to suppress spikes shorter than 50 ns on the input signal, andthe pin is driven by an open drain driver with slew-rate limitation. When this pin is used by the Two-wire Serial Inter-face, the pull-up can still be controlled by the PORTC0 bit.

Table 12-10 and Table 12-11 relate the alternate functions of Port C to the overriding signals shown in Figure 12-5on page 56.

Table 13-10. Overriding Signals for Alternate Functions in PC7:PC4

SignalName PC7/TOSC2 PC6/TOSC1 PC5/TDI PC4/TDO

PUOE AS2 AS2 JTAGEN JTAGEN

PUOV 0 0 1 0

DDOE AS2 AS2 JTAGEN JTAGEN

DDOV 0 0 0 SHIFT_IR + SHIFT_DR

PVOE 0 0 0 JTAGEN

PVOV 0 0 0 TDO

DIEOE AS2 AS2 JTAGEN JTAGEN

DIEOV 0 0 0 0

DI – – – –

AIO T/C2 OSC OUTPUT T/C2 OSC INPUT TDI –


Note: 1. When enabled, the Two-wire Serial Interface enables slew-rate controls on the output pins PC0 and PC1. This is not shown in the figure. In addition, spike filters are connected between the AIO outputs shown in the port figure and the digital logic of the TWI module.

13.3.4 Alternate Functions of Port DThe Port D pins with alternate functions are shown in Table 12-12.


• OC2 – Port D, Bit 7OC2, Timer/Counter2 Output Compare Match output: The PD7 pin can serve as an external output for theTimer/Counter2 Output Compare. The pin has to be configured as an output (DDD7 set (one)) to serve this func-tion. The OC2 pin is also the output pin for the PWM mode timer function.

• ICP1 – Port D, Bit 6ICP1 – Input Capture Pin: The PD6 pin can act as an Input Capture pin for Timer/Counter1.

Table 13-11. Overriding Signals for Alternate Functions in PC3:PC0(1)

Signal Name PC3/TMS PC2/TCK PC1/SDA PC0/SCL

PUOE JTAGEN JTAGEN TWEN TWEN

PUOV 1 1 PORTC1 • PUD PORTC0 • PUD

DDOE JTAGEN JTAGEN TWEN TWEN

DDOV 0 0 SDA_OUT SCL_OUT

PVOE 0 0 TWEN TWEN

PVOV 0 0 0 0

DIEOE JTAGEN JTAGEN 0 0

DIEOV 0 0 0 0

DI – – – –

AIO TMS TCK SDA INPUT SCL INPUT

Table 13-12. Port D Pins Alternate Functions


PD7 OC2 (Timer/Counter2 Output Compare Match Output)

PD6 ICP1 (Timer/Counter1 Input Capture Pin)

PD5 OC1A (Timer/Counter1 Output Compare A Match Output)

PD4 OC1B (Timer/Counter1 Output Compare B Match Output)

PD3 INT1 (External Interrupt 1 Input)

PD2 INT0 (External Interrupt 0 Input)

PD1 TXD (USART Output Pin)

PD0 RXD (USART Input Pin)


• OC1A – Port D, Bit 5OC1A, Output Compare Match A output: The PD5 pin can serve as an external output for the Timer/Counter1 Out-put Compare A. The pin has to be configured as an output (DDD5 set (one)) to serve this function. The OC1A pinis also the output pin for the PWM mode timer function.

• OC1B – Port D, Bit 4OC1B, Output Compare Match B output: The PD4 pin can serve as an external output for the Timer/Counter1 Out-put Compare B. The pin has to be configured as an output (DDD4 set (one)) to serve this function. The OC1B pinis also the output pin for the PWM mode timer function.

• INT1 – Port D, Bit 3INT1, External Interrupt Source 1: The PD3 pin can serve as an external interrupt source.

• INT0 – Port D, Bit 2INT0, External Interrupt Source 0: The PD2 pin can serve as an external interrupt source.

• TXD – Port D, Bit 1TXD, Transmit Data (Data output pin for the USART). When the USART Transmitter is enabled, this pin is config-ured as an output regardless of the value of DDD1.

• RXD – Port D, Bit 0RXD, Receive Data (Data input pin for the USART). When the USART Receiver is enabled this pin is configured asan input regardless of the value of DDD0. When the USART forces this pin to be an input, the pull-up can still becontrolled by the PORTD0 bit.

Table 12-13 and Table 12-14 relate the alternate functions of Port D to the overriding signals shown in Figure 12-5on page 56.

Table 13-13. Overriding Signals for Alternate Functions PD7:PD4

Signal Name PD7/OC2 PD6/ICP1 PD5/OC1A PD4/OC1B

PUOE 0 0 0 0

PUOV 0 0 0 0

DDOE 0 0 0 0

DDOV 0 0 0 0

PVOE OC2 ENABLE 0 OC1A ENABLE OC1B ENABLE

PVOV OC2 0 OC1A OC1B

DIEOE 0 0 0 0

DIEOV 0 0 0 0

DI – ICP1 INPUT – –

AIO – – – –



13.4.1 SFIOR – Special Function I/O Register

• Bit 2 – PUD: Pull-up disableWhen this bit is written to one, the pull-ups in the I/O ports are disabled even if the DDxn and PORTxn Registersare configured to enable the pull-ups ({DDxn, PORTxn} = 0b01). See “Configuring the Pin” on page 51 for moredetails about this feature.

13.4.2 PORTA – Port A Data Register

13.4.3 DDRA – Port A Data Direction Register

13.4.4 PINA – Port A Input Pins Address

Table 13-14. Overriding Signals for Alternate Functions in PD3:PD0

Signal Name PD3/INT1 PD2/INT0 PD1/TXD PD0/RXD

PUOE 0 0 TXEN RXEN

PUOV 0 0 0 PORTD0 • PUD

DDOE 0 0 TXEN RXEN

DDOV 0 0 1 0

PVOE 0 0 TXEN 0

PVOV 0 0 TXD 0

DIEOE INT1 ENABLE INT0 ENABLE 0 0

DIEOV 1 1 0 0

DI INT1 INPUT INT0 INPUT – RXD

AIO – – – –

Bit 7 6 5 4 3 2 1 0

ADTS2 ADTS1 ADTS0 – ACME PUD PSR2 PSR10 SFIOR

Read/Write R/W R/W R/W R R/W R/W R/W R/W


Bit 7 6 5 4 3 2 1 0

PORTA7 PORTA6 PORTA5 PORTA4 PORTA3 PORTA2 PORTA1 PORTA0 PORTA



Bit 7 6 5 4 3 2 1 0

DDA7 DDA6 DDA5 DDA4 DDA3 DDA2 DDA1 DDA0 DDRA



Bit 7 6 5 4 3 2 1 0

PINA7 PINA6 PINA5 PINA4 PINA3 PINA2 PINA1 PINA0 PINA

Read/Write R R R R R R R R

Initial Value N/A N/A N/A N/A N/A N/A N/A N/A


13.4.5 PORTB – Port B Data Register

13.4.6 DDRB – Port B Data Direction Register

13.4.7 PINB – Port B Input Pins Address

13.4.8 PORTC – Port C Data Register

13.4.9 DDRC – Port C Data Direction Register

13.4.10 PINC – Port C Input Pins Address

13.4.11 PORTD – Port D Data Register

13.4.12 DDRD – Port D Data Direction Register

Bit 7 6 5 4 3 2 1 0

PORTB7 PORTB6 PORTB5 PORTB4 PORTB3 PORTB2 PORTB1 PORTB0 PORTB



Bit 7 6 5 4 3 2 1 0

DDB7 DDB6 DDB5 DDB4 DDB3 DDB2 DDB1 DDB0 DDRB



Bit 7 6 5 4 3 2 1 0

PINB7 PINB6 PINB5 PINB4 PINB3 PINB2 PINB1 PINB0 PINB



Bit 7 6 5 4 3 2 1 0

PORTC7 PORTC6 PORTC5 PORTC4 PORTC3 PORTC2 PORTC1 PORTC0 PORTC



Bit 7 6 5 4 3 2 1 0

DDC7 DDC6 DDC5 DDC4 DDC3 DDC2 DDC1 DDC0 DDRC



Bit 7 6 5 4 3 2 1 0

PINC7 PINC6 PINC5 PINC4 PINC3 PINC2 PINC1 PINC0 PINC



Bit 7 6 5 4 3 2 1 0

PORTD7 PORTD6 PORTD5 PORTD4 PORTD3 PORTD2 PORTD1 PORTD0 PORTD



Bit 7 6 5 4 3 2 1 0

DDD7 DDD6 DDD5 DDD4 DDD3 DDD2 DDD1 DDD0 DDRD




13.4.13 PIND – Port D Input Pins Address

Bit 7 6 5 4 3 2 1 0

PIND7 PIND6 PIND5 PIND4 PIND3 PIND2 PIND1 PIND0 PIND




14. External InterruptsThe External Interrupts are triggered by the INT0, INT1, and INT2 pins. Observe that, if enabled, the interrupts willtrigger even if the INT0:2 pins are configured as outputs. This feature provides a way of generating a softwareinterrupt. The external interrupts can be triggered by a falling or rising edge or a low level (INT2 is only an edge trig-gered interrupt). This is set up as indicated in the specification for the MCU Control Register – MCUCR – and MCUControl and Status Register – MCUCSR. When the external interrupt is enabled and is configured as level trig-gered (only INT0/INT1), the interrupt will trigger as long as the pin is held low. Note that recognition of falling orrising edge interrupts on INT0 and INT1 requires the presence of an I/O clock, described in “Clock Systems andtheir Distribution” on page 25. Low level interrupts on INT0/INT1 and the edge interrupt on INT2 are detected asyn-chronously. This implies that these interrupts can be used for waking the part also from sleep modes other thanIdle mode. The I/O clock is halted in all sleep modes except Idle mode.

Note that if a level triggered interrupt is used for wake-up from Power-down mode, the changed level must be heldfor some time to wake up the MCU. This makes the MCU less sensitive to noise. The changed level is sampledtwice by the Watchdog Oscillator clock. The period of the Watchdog Oscillator is 1 µs (nominal) at 5.0V and 25C.The frequency of the Watchdog Oscillator is voltage dependent as shown in “Electrical Characteristics” on page296. The MCU will wake up if the input has the required level during this sampling or if it is held until the end of thestart-up time. The start-up time is defined by the SUT fuses as described in “System Clock and Clock Options” onpage 25. If the level is sampled twice by the Watchdog Oscillator clock but disappears before the end of the start-up time, the MCU will still wake up, but no interrupt will be generated. The required level must be held long enoughfor the MCU to complete the wake up to trigger the level interrupt.


14.1.1 MCUCR – MCU Control RegisterThe MCU Control Register contains control bits for interrupt sense control and general MCU functions.

• Bit 3, 2 – ISC11, ISC10: Interrupt Sense Control 1 Bit 1 and Bit 0The External Interrupt 1 is activated by the external pin INT1 if the SREG I-bit and the corresponding interruptmask in the GICR are set. The level and edges on the external INT1 pin that activate the interrupt are defined inTable 13-1. The value on the INT1 pin is sampled before detecting edges. If edge or toggle interrupt is selected,pulses that last longer than one clock period will generate an interrupt. Shorter pulses are not guaranteed to gener-ate an interrupt. If low level interrupt is selected, the low level must be held until the completion of the currentlyexecuting instruction to generate an interrupt.

• Bit 1, 0 – ISC01, ISC00: Interrupt Sense Control 0 Bit 1 and Bit 0

Bit 7 6 5 4 3 2 1 0

SE SM2 SM1 SM0 ISC11 ISC10 ISC01 ISC00 MCUCR



Table 14-1. Interrupt 1 Sense Control

ISC11 ISC10 Description

0 0 The low level of INT1 generates an interrupt request.

0 1 Any logical change on INT1 generates an interrupt request.

1 0 The falling edge of INT1 generates an interrupt request.

1 1 The rising edge of INT1 generates an interrupt request.


The External Interrupt 0 is activated by the external pin INT0 if the SREG I-flag and the corresponding interruptmask are set. The level and edges on the external INT0 pin that activate the interrupt are defined in Table 13-2.The value on the INT0 pin is sampled before detecting edges. If edge or toggle interrupt is selected, pulses that lastlonger than one clock period will generate an interrupt. Shorter pulses are not guaranteed to generate an interrupt.If low level interrupt is selected, the low level must be held until the completion of the currently executing instructionto generate an interrupt.

14.1.2 MCUCSR – MCU Control and Status Register

• Bit 6 – ISC2: Interrupt Sense Control 2The Asynchronous External Interrupt 2 is activated by the external pin INT2 if the SREG I-bit and the correspond-ing interrupt mask in GICR are set. If ISC2 is written to zero, a falling edge on INT2 activates the interrupt. If ISC2is written to one, a rising edge on INT2 activates the interrupt. Edges on INT2 are registered asynchronously.Pulses on INT2 wider than the minimum pulse width given in Table 13-3 will generate an interrupt. Shorter pulsesare not guaranteed to generate an interrupt. When changing the ISC2 bit, an interrupt can occur. Therefore, it isrecommended to first disable INT2 by clearing its Interrupt Enable bit in the GICR Register. Then, the ISC2 bit canbe changed. Finally, the INT2 Interrupt Flag should be cleared by writing a logical one to its Interrupt Flag bit(INTF2) in the GIFR Register before the interrupt is re-enabled.

14.1.3 GICR – General Interrupt Control Register

• Bit 7 – INT1: External Interrupt Request 1 EnableWhen the INT1 bit is set (one) and the I-bit in the Status Register (SREG) is set (one), the external pin interrupt isenabled. The Interrupt Sense Control1 bits 1/0 (ISC11 and ISC10) in the MCU General Control Register (MCUCR)define whether the External Interrupt is activated on rising and/or falling edge of the INT1 pin or level sensed. Activ-ity on the pin will cause an interrupt request even if INT1 is configured as an output. The corresponding interrupt ofExternal Interrupt Request 1 is executed from the INT1 interrupt Vector.

Table 14-2. Interrupt 0 Sense Control

ISC01 ISC00 Description

0 0 The low level of INT0 generates an interrupt request.

0 1 Any logical change on INT0 generates an interrupt request.

1 0 The falling edge of INT0 generates an interrupt request.

1 1 The rising edge of INT0 generates an interrupt request.

Bit 7 6 5 4 3 2 1 0




Table 14-3. Asynchronous External Interrupt Characteristics

Symbol Parameter Condition Min Typ Max Units

tINTMinimum pulse width for asynchronous external interrupt

50 ns

Bit 7 6 5 4 3 2 1 0

INT1 INT0 INT2 – – – IVSEL IVCE GICR

Read/Write R/W R/W R/W R R R R/W R/W



• Bit 6 – INT0: External Interrupt Request 0 EnableWhen the INT0 bit is set (one) and the I-bit in the Status Register (SREG) is set (one), the external pin interrupt isenabled. The Interrupt Sense Control0 bits 1/0 (ISC01 and ISC00) in the MCU General Control Register (MCUCR)define whether the External Interrupt is activated on rising and/or falling edge of the INT0 pin or level sensed. Activ-ity on the pin will cause an interrupt request even if INT0 is configured as an output. The corresponding interrupt ofExternal Interrupt Request 0 is executed from the INT0 interrupt vector.

• Bit 5 – INT2: External Interrupt Request 2 EnableWhen the INT2 bit is set (one) and the I-bit in the Status Register (SREG) is set (one), the external pin interrupt isenabled. The Interrupt Sense Control2 bit (ISC2) in the MCU Control and Status Register (MCUCSR) defineswhether the External Interrupt is activated on rising or falling edge of the INT2 pin. Activity on the pin will cause aninterrupt request even if INT2 is configured as an output. The corresponding interrupt of External Interrupt Request2 is executed from the INT2 Interrupt Vector.

14.1.4 GIFR – General Interrupt Flag Register

• Bit 7 – INTF1: External Interrupt Flag 1When an edge or logic change on the INT1 pin triggers an interrupt request, INTF1 becomes set (one). If the I-bit inSREG and the INT1 bit in GICR are set (one), the MCU will jump to the corresponding Interrupt Vector. The flag iscleared when the interrupt routine is executed. Alternatively, the flag can be cleared by writing a logical one to it.This flag is always cleared when INT1 is configured as a level interrupt.

• Bit 6 – INTF0: External Interrupt Flag 0When an edge or logic change on the INT0 pin triggers an interrupt request, INTF0 becomes set (one). If the I-bit inSREG and the INT0 bit in GICR are set (one), the MCU will jump to the corresponding interrupt vector. The flag iscleared when the interrupt routine is executed. Alternatively, the flag can be cleared by writing a logical one to it.This flag is always cleared when INT0 is configured as a level interrupt.

• Bit 5 – INTF2: External Interrupt Flag 2When an event on the INT2 pin triggers an interrupt request, INTF2 becomes set (one). If the I-bit in SREG and theINT2 bit in GICR are set (one), the MCU will jump to the corresponding Interrupt Vector. The flag is cleared whenthe interrupt routine is executed. Alternatively, the flag can be cleared by writing a logical one to it. Note that whenentering some sleep modes with the INT2 interrupt disabled, the input buffer on this pin will be disabled. This maycause a logic change in internal signals which will set the INTF2 Flag. See “Digital Input Enable and Sleep Modes”on page 54 for more information.

Bit 7 6 5 4 3 2 1 0

INTF1 INTF0 INTF2 – – – – – GIFR

Read/Write R/W R/W R/W R R R R R



15. 8-bit Timer/Counter0 with PWM

15.1 Features• Single Compare Unit Counter• Clear Timer on Compare Match (Auto Reload)• Glitch-free, Phase Correct Pulse Width Modulator (PWM)• Frequency Generator• External Event Counter• 10-bit Clock Prescaler• Overflow and Compare Match Interrupt Sources (TOV0 and OCF0)

15.2 OverviewTimer/Counter0 is a general purpose, single compare unit, 8-bit Timer/Counter module. A simplified block diagramof the 8-bit Timer/Counter is shown in Figure 14-1. For the actual placement of I/O pins, refer to “PinoutATmega32A” on page 2. CPU accessible I/O Registers, including I/O bits and I/O pins, are shown in bold. Thedevice-specific I/O Register and bit locations are listed in the “Register Description” on page 84.

Figure 15-1. 8-bit Timer/Counter Block Diagram

15.2.1 RegistersThe Timer/Counter (TCNT0) and Output Compare Register (OCR0) are 8-bit registers. Interrupt request (abbrevi-ated to Int.Req. in the figure) signals are all visible in the Timer Interrupt Flag Register (TIFR). All interrupts areindividually masked with the Timer Interrupt Mask Register (TIMSK). TIFR and TIMSK are not shown in the figuresince these registers are shared by other timer units.

The Timer/Counter can be clocked internally, via the prescaler, or by an external clock source on the T0 pin. TheClock Select logic block controls which clock source and edge the Timer/Counter uses to increment (or decrement)

Timer/Counter

DAT

AB

US

=

TCNTn

WaveformGeneration

OCn

= 0

Control Logic

= 0xFF

BOTTOM

count

clear

direction

TOVn(Int.Req.)

OCRn

TCCRn

Clock Select

TnEdge

Detector

( From Prescaler )

clkTn

TOP

OCn(Int.Req.)


its value. The Timer/Counter is inactive when no clock source is selected. The output from the Clock Select logic isreferred to as the timer clock (clkT0).

The double buffered Output Compare Register (OCR0) is compared with the Timer/Counter value at all times. Theresult of the compare can be used by the waveform generator to generate a PWM or variable frequency output onthe Output Compare Pin (OC0). See “Output Compare Unit” on page 75. for details. The compare match event willalso set the Compare Flag (OCF0) which can be used to generate an output compare interrupt request.

15.2.2 DefinitionsMany register and bit references in this document are written in general form. A lower case “n” replaces theTimer/Counter number, in this case 0. However, when using the register or bit defines in a program, the preciseform must be used, that is, TCNT0 for accessing Timer/Counter0 counter value and so on.

The definitions in Table 14-1 are also used extensively throughout the document.

15.3 Timer/Counter Clock SourcesThe Timer/Counter can be clocked by an internal or an external clock source. The clock source is selected by theclock select logic which is controlled by the clock select (CS02:0) bits located in the Timer/Counter Control Regis-ter (TCCR0). For details on clock sources and prescaler, see “Timer/Counter0 and Timer/Counter1 Prescalers” onpage 88.

15.4 Counter UnitThe main part of the 8-bit Timer/Counter is the programmable bi-directional counter unit. Figure 14-2 shows ablock diagram of the counter and its surroundings.

Figure 15-2. Counter Unit Block Diagram

Signal description (internal signals):

count Increment or decrement TCNT0 by 1.

direction Select between increment and decrement.

clear Clear TCNT0 (set all bits to zero).

clkTn Timer/Counter clock, referred to as clkT0 in the following.

Table 15-1. Definitions

BOTTOM The counter reaches the BOTTOM when it becomes 0x00.

MAX The counter reaches its MAXimum when it becomes 0xFF (decimal 255).

TOP The counter reaches the TOP when it becomes equal to the highest value in thecount sequence. The TOP value can be assigned to be the fixed value 0xFF(MAX) or the value stored in the OCR0 Register. The assignment is dependenton the mode of operation.

DATA BUS

TCNTn Control Logic

count

TOVn(Int. Req.)

Clock Select

TOP

TnEdge

Detector

( From Prescaler )

clkTn

BOTTOM

direction

clear


TOP Signalize that TCNT0 has reached maximum value.

BOTTOM Signalize that TCNT0 has reached minimum value (zero).

Depending of the mode of operation used, the counter is cleared, incremented, or decremented at each timer clock(clkT0). clkT0 can be generated from an external or internal clock source, selected by the Clock Select bits(CS02:0). When no clock source is selected (CS02:0 = 0) the timer is stopped. However, the TCNT0 value can beaccessed by the CPU, regardless of whether clkT0 is present or not. A CPU write overrides (has priority over) allcounter clear or count operations.

The counting sequence is determined by the setting of the WGM01 and WGM00 bits located in the Timer/CounterControl Register (TCCR0). There are close connections between how the counter behaves (counts) and howwaveforms are generated on the Output Compare output OC0. For more details about advanced countingsequences and waveform generation, see “Modes of Operation” on page 78.

The Timer/Counter Overflow (TOV0) Flag is set according to the mode of operation selected by the WGM01:0 bits.TOV0 can be used for generating a CPU interrupt.

15.5 Output Compare UnitThe 8-bit comparator continuously compares TCNT0 with the Output Compare Register (OCR0). WheneverTCNT0 equals OCR0, the comparator signals a match. A match will set the Output Compare Flag (OCF0) at thenext timer clock cycle. If enabled (OCIE0 = 1 and Global Interrupt Flag in SREG is set), the Output Compare Flaggenerates an output compare interrupt. The OCF0 Flag is automatically cleared when the interrupt is executed.Alternatively, the OCF0 Flag can be cleared by software by writing a logical one to its I/O bit location. The wave-form generator uses the match signal to generate an output according to operating mode set by the WGM01:0 bitsand Compare Output mode (COM01:0) bits. The max and bottom signals are used by the waveform generator forhandling the special cases of the extreme values in some modes of operation (See “Modes of Operation” on page78.).

Figure 14-3 shows a block diagram of the output compare unit.

Figure 15-3. Output Compare Unit, Block Diagram

OCFn (Int.Req.)

= (8-bit Comparator )

OCRn

OCn

DATA BUS

TCNTn

WGMn1:0

Waveform Generator

top

FOCn

COMn1:0

bottom


The OCR0 Register is double buffered when using any of the Pulse Width Modulation (PWM) modes. For the nor-mal and Clear Timer on Compare (CTC) modes of operation, the double buffering is disabled. The double bufferingsynchronizes the update of the OCR0 Compare Register to either top or bottom of the counting sequence. Thesynchronization prevents the occurrence of odd-length, non-symmetrical PWM pulses, thereby making the outputglitch-free.

The OCR0 Register access may seem complex, but this is not case. When the double buffering is enabled, theCPU has access to the OCR0 Buffer Register, and if double buffering is disabled the CPU will access the OCR0directly.

15.5.1 Force Output CompareIn non-PWM waveform generation modes, the match output of the comparator can be forced by writing a one tothe Force Output Compare (FOC0) bit. Forcing compare match will not set the OCF0 Flag or reload/clear the timer,but the OC0 pin will be updated as if a real compare match had occurred (the COM0[1:0] bits settings definewhether the OC0 pin is set, cleared or toggled).

15.5.2 Compare Match Blocking by TCNT0 WriteAll CPU write operations to the TCNT0 Register will block any compare match that occur in the next timer clockcycle, even when the timer is stopped. This feature allows OCR0 to be initialized to the same value as TCNT0 with-out triggering an interrupt when the Timer/Counter clock is enabled.

15.5.3 Using the Output Compare UnitSince writing TCNT0 in any mode of operation will block all compare matches for one timer clock cycle, there arerisks involved when changing TCNT0 when using the output compare unit, independently of whether theTimer/Counter is running or not. If the value written to TCNT0 equals the OCR0 value, the compare match will bemissed, resulting in incorrect waveform generation. Similarly, do not write the TCNT0 value equal to BOTTOMwhen the counter is downcounting.

The setup of the OC0 should be performed before setting the Data Direction Register for the port pin to output. Theeasiest way of setting the OC0 value is to use the Force Output Compare (FOC0) strobe bits in Normal mode. TheOC0 Register keeps its value even when changing between waveform generation modes.

Be aware that the COM0[1:0] bits are not double buffered together with the compare value. Changing theCOM0[1:0] bits will take effect immediately.

15.6 Compare Match Output UnitThe Compare Output mode (COM0[1:0]) bits have two functions. The Waveform Generator uses the COM0[1:0]bits for defining the Output Compare (OC0) state at the next compare match. Also, the COM0[1:0] bits control theOC0 pin output source. Figure 14-4 shows a simplified schematic of the logic affected by the COM0[1:0] bit setting.The I/O Registers, I/O bits, and I/O pins in the figure are shown in bold. Only the parts of the general I/O port Con-trol Registers (DDR and PORT) that are affected by the COM0[1:0] bits are shown. When referring to the OC0state, the reference is for the internal OC0 Register, not the OC0 pin. If a System Reset occur, the OC0 Register isreset to “0”.


Figure 15-4. Compare Match Output Unit, Schematic

The general I/O port function is overridden by the Output Compare (OC0) from the Waveform Generator if either ofthe COM0[1:0] bits are set. However, the OC0 pin direction (input or output) is still controlled by the Data DirectionRegister (DDR) for the port pin. The Data Direction Register bit for the OC0 pin (DDR_OC0) must be set as outputbefore the OC0 value is visible on the pin. The port override function is independent of the Waveform Generationmode.

The design of the output compare pin logic allows initialization of the OC0 state before the output is enabled. Notethat some COM01:0 bit settings are reserved for certain modes of operation. See “Register Description” on page84.

15.6.1 Compare Output Mode and Waveform GenerationThe Waveform Generator uses the COM0[1:0] bits differently in normal, CTC, and PWM modes. For all modes,setting the COM0[1:0] = 0 tells the waveform generator that no action on the OC0 Register is to be performed onthe next compare match. For compare output actions in the non-PWM modes refer to Table 14-3 on page 85. Forfast PWM mode, refer to Table 14-4 on page 85, and for phase correct PWM refer to Table 14-5 on page 86.

A change of the COM0[1:0] bits state will have effect at the first compare match after the bits are written. For non-PWM modes, the action can be forced to have immediate effect by using the FOC0 strobe bits.

15.7 Modes of OperationThe mode of operation, that is, the behavior of the Timer/Counter and the Output Compare pins, is defined by thecombination of the Waveform Generation mode (WGM01:0) and Compare Output mode (COM0[1:0]) bits. TheCompare Output mode bits do not affect the counting sequence, while the Waveform Generation mode bits do.The COM0[1:0] bits control whether the PWM output generated should be inverted or not (inverted or non-invertedPWM). For non-PWM modes the COM0[1:0] bits control whether the output should be set, cleared, or toggled at acompare match (See “Compare Match Output Unit” on page 77.).

For detailed timing information refer to Figure 14-8, Figure 14-9, Figure 14-10 and Figure 14-11 in “Timer/CounterTiming Diagrams” on page 82.

15.7.1 Normal ModeThe simplest mode of operation is the normal mode (WGM01:0 = 0). In this mode the counting direction is alwaysup (incrementing), and no counter clear is performed. The counter simply overruns when it passes its maximum 8-bit value (TOP = 0xFF) and then restarts from the bottom (0x00). In normal operation the Timer/Counter Overflow

PORT

DDR

D Q

D Q

OCnPinOCn

D QWaveformGenerator

COMn1

COMn0

0

1

DAT

A B

US

FOCn

clkI/O


Flag (TOV0) will be set in the same timer clock cycle as the TCNT0 becomes zero. The TOV0 Flag in this casebehaves like a ninth bit, except that it is only set, not cleared. However, combined with the timer overflow interruptthat automatically clears the TOV0 Flag, the timer resolution can be increased by software. There are no specialcases to consider in the normal mode, a new counter value can be written anytime.

The output compare unit can be used to generate interrupts at some given time. Using the output compare to gen-erate waveforms in Normal mode is not recommended, since this will occupy too much of the CPU time.

15.7.2 Clear Timer on Compare Match (CTC) ModeIn Clear Timer on Compare or CTC mode (WGM01:0 = 2), the OCR0 Register is used to manipulate the counterresolution. In CTC mode the counter is cleared to zero when the counter value (TCNT0) matches the OCR0. TheOCR0 defines the top value for the counter, hence also its resolution. This mode allows greater control of the com-pare match output frequency. It also simplifies the operation of counting external events.

The timing diagram for the CTC mode is shown in Figure 14-5. The counter value (TCNT0) increases until a com-pare match occurs between TCNT0 and OCR0, and then counter (TCNT0) is cleared.

Figure 15-5. CTC Mode, Timing Diagram

An interrupt can be generated each time the counter value reaches the TOP value by using the OCF0 Flag. If theinterrupt is enabled, the interrupt handler routine can be used for updating the TOP value. However, changing TOPto a value close to BOTTOM when the counter is running with none or a low prescaler value must be done withcare since the CTC mode does not have the double buffering feature. If the new value written to OCR0 is lowerthan the current value of TCNT0, the counter will miss the compare match. The counter will then have to count toits maximum value (0xFF) and wrap around starting at 0x00 before the compare match can occur.

For generating a waveform output in CTC mode, the OC0 output can be set to toggle its logical level on each com-pare match by setting the Compare Output mode bits to toggle mode (COM01:0 = 1). The OC0 value will not bevisible on the port pin unless the data direction for the pin is set to output. The waveform generated will have amaximum frequency of fOC0 = fclk_I/O/2 when OCR0 is set to zero (0x00). The waveform frequency is defined by thefollowing equation:

The N variable represents the prescale factor (1, 8, 64, 256, or 1024).

As for the Normal mode of operation, the TOV0 Flag is set in the same timer clock cycle that the counter countsfrom MAX to 0x00.

TCNTn

OCn(Toggle)

OCn Interrupt Flag Set

1 4Period 2 3

(COMn1:0 = 1)

fOCn

fclk_I/O

2 N 1 OCRn+ -----------------------------------------------=


15.7.3 Fast PWM ModeThe fast Pulse Width Modulation or fast PWM mode (WGM01:0 = 3) provides a high frequency PWM waveformgeneration option. The fast PWM differs from the other PWM option by its single-slope operation. The countercounts from BOTTOM to MAX then restarts from BOTTOM. In non-inverting Compare Output mode, the OutputCompare (OC0) is cleared on the compare match between TCNT0 and OCR0, and set at BOTTOM. In invertingCompare Output mode, the output is set on compare match and cleared at BOTTOM. Due to the single-slope oper-ation, the operating frequency of the fast PWM mode can be twice as high as the phase correct PWM mode thatuse dual-slope operation. This high frequency makes the fast PWM mode well suited for power regulation, rectifi-cation, and DAC applications. High frequency allows physically small sized external components (coils,capacitors), and therefore reduces total system cost.

In fast PWM mode, the counter is incremented until the counter value matches the MAX value. The counter is thencleared at the following timer clock cycle. The timing diagram for the fast PWM mode is shown in Figure 14-6. TheTCNT0 value is in the timing diagram shown as a histogram for illustrating the single-slope operation. The diagramincludes non-inverted and inverted PWM outputs. The small horizontal line marks on the TCNT0 slopes representcompare matches between OCR0 and TCNT0.

Figure 15-6. Fast PWM Mode, Timing Diagram

The Timer/Counter Overflow Flag (TOV0) is set each time the counter reaches MAX. If the interrupt is enabled, theinterrupt handler routine can be used for updating the compare value.

In fast PWM mode, the compare unit allows generation of PWM waveforms on the OC0 pin. Setting the COM01:0bits to 2 will produce a non-inverted PWM and an inverted PWM output can be generated by setting the COM01:0to 3 (See Table 14-4 on page 85). The actual OC0 value will only be visible on the port pin if the data direction forthe port pin is set as output. The PWM waveform is generated by setting (or clearing) the OC0 Register at the com-pare match between OCR0 and TCNT0, and clearing (or setting) the OC0 Register at the timer clock cycle thecounter is cleared (changes from MAX to BOTTOM).

The PWM frequency for the output can be calculated by the following equation:


The extreme values for the OCR0 Register represents special cases when generating a PWM waveform output inthe fast PWM mode. If the OCR0 is set equal to BOTTOM, the output will be a narrow spike for each MAX+1 timer

TCNTn

OCRn Update andTOVn Interrupt Flag Set

1Period 2 3

OCn

OCn

(COMn1:0 = 2)

(COMn1:0 = 3)

OCRn Interrupt Flag Set

4 5 6 7

fOCnPWM

fclk_I/O

N 256------------------=


clock cycle. Setting the OCR0 equal to MAX will result in a constantly high or low output (depending on the polarityof the output set by the COM01:0 bits.)

15.7.4 Phase Correct PWM ModeThe phase correct PWM mode (WGM0[1:0] = 1) provides a high resolution phase correct PWM waveform genera-tion option. The phase correct PWM mode is based on a dual-slope operation. The counter counts repeatedly fromBOTTOM to MAX and then from MAX to BOTTOM. In non-inverting Compare Output mode, the Output Compare(OC0) is cleared on the compare match between TCNT0 and OCR0 while upcounting, and set on the comparematch while downcounting. In inverting Output Compare mode, the operation is inverted. The dual-slope operationhas lower maximum operation frequency than single slope operation. However, due to the symmetric feature of thedual-slope PWM modes, these modes are preferred for motor control applications.

The PWM resolution for the phase correct PWM mode is fixed to eight bits. In phase correct PWM mode the coun-ter is incremented until the counter value matches MAX. When the counter reaches MAX, it changes the countdirection. The TCNT0 value will be equal to MAX for one timer clock cycle. The timing diagram for the phase cor-rect PWM mode is shown on Figure 14-7. The TCNT0 value is in the timing diagram shown as a histogram forillustrating the dual-slope operation. The diagram includes non-inverted and inverted PWM outputs. The small hor-izontal line marks on the TCNT0 slopes represent compare matches between OCR0 and TCNT0.

Figure 15-7. Phase Correct PWM Mode, Timing Diagram

The Timer/Counter Overflow Flag (TOV0) is set each time the counter reaches BOTTOM. The Interrupt Flag canbe used to generate an interrupt each time the counter reaches the BOTTOM value.

In phase correct PWM mode, the compare unit allows generation of PWM waveforms on the OC0 pin. Setting theCOM01:0 bits to 2 will produce a non-inverted PWM. An inverted PWM output can be generated by setting theCOM01:0 to 3 (see Table 14-5 on page 86). The actual OC0 value will only be visible on the port pin if the datadirection for the port pin is set as output. The PWM waveform is generated by clearing (or setting) the OC0 Regis-ter at the compare match between OCR0 and TCNT0 when the counter increments, and setting (or clearing) the

TOVn Interrupt Flag Set


1 2 3

TCNTn

Period

OCn

OCn

(COMn1:0 = 2)

(COMn1:0 = 3)

OCRn Update


OC0 Register at compare match between OCR0 and TCNT0 when the counter decrements. The PWM frequencyfor the output when using phase correct PWM can be calculated by the following equation:


The extreme values for the OCR0 Register represent special cases when generating a PWM waveform output inthe phase correct PWM mode. If the OCR0 is set equal to BOTTOM, the output will be continuously low and if setequal to MAX the output will be continuously high for non-inverted PWM mode. For inverted PWM the output willhave the opposite logic values.

At the very start of period 2 in Figure 14-7 OCn has a transition from high to low even though there is no CompareMatch. The point of this transition is to guarantee symmetry around BOTTOM. There are two cases that give atransition without Compare Match:

• OCR0A changes its value from MAX, like in Figure 14-7. When the OCR0A value is MAX the OCn pin value is the same as the result of a down-counting Compare Match. To ensure symmetry around BOTTOM the OCn value at MAX must correspond to the result of an up-counting Compare Match.

• The timer starts counting from a value higher than the one in OCR0A, and for that reason misses the Compare Match and hence the OCn change that would have happened on the way up.

15.8 Timer/Counter Timing DiagramsThe Timer/Counter is a synchronous design and the timer clock (clkT0) is therefore shown as a clock enable signalin the following figures. The figures include information on when Interrupt Flags are set. Figure 14-8 contains timingdata for basic Timer/Counter operation. The figure shows the count sequence close to the MAX value in all modesother than phase correct PWM mode.

Figure 15-8. Timer/Counter Timing Diagram, no Prescaling

Figure 14-9 shows the same timing data, but with the prescaler enabled.

fOCnPCPWM

fclk_I/O

N 510------------------=

clkTn(clkI/O/1)

TOVn

clkI/O

TCNTn MAX - 1 MAX BOTTOM BOTTOM + 1


Figure 15-9. Timer/Counter Timing Diagram, with Prescaler (fclk_I/O/8)

Figure 14-10 shows the setting of OCF0 in all modes except CTC mode.

Figure 15-10. Timer/Counter Timing Diagram, Setting of OCF0, with Prescaler (fclk_I/O/8)

Figure 14-11 shows the setting of OCF0 and the clearing of TCNT0 in CTC mode.

TOVn


clkI/O

clkTn(clkI/O/8)

OCFn

OCRn

TCNTn

OCRn Value

OCRn - 1 OCRn OCRn + 1 OCRn + 2

clkI/O

clkTn(clkI/O/8)


Figure 15-11. Timer/Counter Timing Diagram, Clear Timer on Compare Match Mode, with Prescaler (fclk_I/O/8)


15.9.1 TCCR0 – Timer/Counter Control Register

• Bit 7 – FOC0: Force Output CompareThe FOC0 bit is only active when the WGM00 bit specifies a non-PWM mode. However, for ensuring compatibilitywith future devices, this bit must be set to zero when TCCR0 is written when operating in PWM mode. When writ-ing a logical one to the FOC0 bit, an immediate compare match is forced on the Waveform Generation unit. TheOC0 output is changed according to its COM0[1:0] bits setting. Note that the FOC0 bit is implemented as a strobe.Therefore it is the value present in the COM0[1:0] bits that determines the effect of the forced compare.

A FOC0 strobe will not generate any interrupt, nor will it clear the timer in CTC mode using OCR0 as TOP.

The FOC0 bit is always read as zero.

• Bit 6, 3 – WGM0[1:0]: Waveform Generation ModeThese bits control the counting sequence of the counter, the source for the maximum (TOP) counter value, andwhat type of Waveform Generation to be used. Modes of operation supported by the Timer/Counter unit are: Nor-mal mode, Clear Timer on Compare Match (CTC) mode, and two types of Pulse Width Modulation (PWM) modes.See Table 14-2 and “Modes of Operation” on page 78.

OCFn

OCRn

TCNTn(CTC)

TOP

TOP - 1 TOP BOTTOM BOTTOM + 1

clkI/O

clkTn(clkI/O/8)

Bit 7 6 5 4 3 2 1 0

FOC0 WGM00 COM01 COM00 WGM01 CS02 CS01 CS00 TCCR0

Read/Write W R/W R/W R/W R/W R/W R/W R/W


Table 15-2. Waveform Generation Mode Bit Description(1)

ModeWGM01(CTC0)

WGM00(PWM0)

Timer/Counter Mode of Operation TOP

Update ofOCR0

TOV0 FlagSet-on

0 0 0 Normal 0xFF Immediate MAX

1 0 1 PWM, Phase Correct 0xFF TOP BOTTOM

2 1 0 CTC OCR0 Immediate MAX

3 1 1 Fast PWM 0xFF BOTTOM MAX


Note: 1. The CTC0 and PWM0 bit definition names are now obsolete. Use the WGM01:0 definitions. However, the function-ality and location of these bits are compatible with previous versions of the timer.

• Bit 5:4 – COM0[1:0]: Compare Match Output ModeThese bits control the Output Compare pin (OC0) behavior. If one or both of the COM01:0 bits are set, the OC0output overrides the normal port functionality of the I/O pin it is connected to. However, note that the Data DirectionRegister (DDR) bit corresponding to the OC0 pin must be set in order to enable the output driver.

When OC0 is connected to the pin, the function of the COM01:0 bits depends on the WGM01:0 bit setting. Table14-3 shows the COM01:0 bit functionality when the WGM01:0 bits are set to a normal or CTC mode (non-PWM).

Table 14-4 shows the COM01:0 bit functionality when the WGM01:0 bits are set to fast PWM mode.

Note: 1. A special case occurs when OCR0 equals TOP and COM01 is set. In this case, the compare match is ignored, but the set or clear is done at BOTTOM. See “Fast PWM Mode” on page 79 for more details.

Table 14-5 shows the COM0[1:0] bit functionality when the WGM01:0 bits are set to phase correct PWM mode.

Note: 1. A special case occurs when OCR0 equals TOP and COM01 is set. In this case, the compare match is ignored, butthe set or clear is done at TOP. See “Phase Correct PWM Mode” on page 80 for more details.

• Bit 2:0 – CS02:0: Clock Select

Table 15-3. Compare Output Mode, non-PWM Mode

COM01 COM00 Description

0 0 Normal port operation, OC0 disconnected.

0 1 Toggle OC0 on compare match

1 0 Clear OC0 on compare match

1 1 Set OC0 on compare match

Table 15-4. Compare Output Mode, Fast PWM Mode(1)



0 1 Reserved

1 0 Clear OC0 on compare match, set OC0 at BOTTOM,(nin-inverting mode)

1 1 Set OC0 on compare match, clear OC0 at BOTTOM,(inverting mode)

Table 15-5. Compare Output Mode, Phase Correct PWM Mode(1)



0 1 Reserved

1 0 Clear OC0 on compare match when up-counting. Set OC0 on compare match when downcounting.

1 1 Set OC0 on compare match when up-counting. Clear OC0 on compare match when downcounting.


The three Clock Select bits select the clock source to be used by the Timer/Counter.

If external pin modes are used for the Timer/Counter0, transitions on the T0 pin will clock the counter even if thepin is configured as an output. This feature allows software control of the counting.

15.9.2 TCNT0 – Timer/Counter Register

The Timer/Counter Register gives direct access, both for read and write operations, to the Timer/Counter unit 8-bitcounter. Writing to the TCNT0 Register blocks (removes) the compare match on the following timer clock. Modify-ing the counter (TCNT0) while the counter is running, introduces a risk of missing a compare match betweenTCNT0 and the OCR0 Register.

15.9.3 OCR0 – Output Compare Register

The Output Compare Register contains an 8-bit value that is continuously compared with the counter value(TCNT0). A match can be used to generate an output compare interrupt, or to generate a waveform output on theOC0 pin.

15.9.4 TIMSK – Timer/Counter Interrupt Mask Register

• Bit 1 – OCIE0: Timer/Counter0 Output Compare Match Interrupt EnableWhen the OCIE0 bit is written to one, and the I-bit in the Status Register is set (one), the Timer/Counter0 CompareMatch interrupt is enabled. The corresponding interrupt is executed if a compare match in Timer/Counter0 occurs,that is, when the OCF0 bit is set in the Timer/Counter Interrupt Flag Register – TIFR.

• Bit 0 – TOIE0: Timer/Counter0 Overflow Interrupt Enable

Table 15-6. Clock Select Bit Description

CS02 CS01 CS00 Description

0 0 0 No clock source (Timer/Counter stopped).

0 0 1 clkI/O/(No prescaling)

0 1 0 clkI/O/8 (From prescaler)




1 1 0 External clock source on T0 pin. Clock on falling edge.

1 1 1 External clock source on T0 pin. Clock on rising edge.

Bit 7 6 5 4 3 2 1 0

TCNT0[7:0] TCNT0



Bit 7 6 5 4 3 2 1 0

OCR0[7:0] OCR0



Bit 7 6 5 4 3 2 1 0

OCIE2 TOIE2 TICIE1 OCIE1A OCIE1B TOIE1 OCIE0 TOIE0 TIMSK




When the TOIE0 bit is written to one, and the I-bit in the Status Register is set (one), the Timer/Counter0 Overflowinterrupt is enabled. The corresponding interrupt is executed if an overflow in Timer/Counter0 occurs, that is, whenthe TOV0 bit is set in the Timer/Counter Interrupt Flag Register – TIFR.

15.9.5 TIFR – Timer/Counter Interrupt Flag Register

• Bit 1 – OCF0: Output Compare Flag 0The OCF0 bit is set (one) when a compare match occurs between the Timer/Counter0 and the data in OCR0 –Output Compare Register0. OCF0 is cleared by hardware when executing the corresponding interrupt handlingvector. Alternatively, OCF0 is cleared by writing a logic one to the flag. When the I-bit in SREG, OCIE0(Timer/Counter0 Compare Match Interrupt Enable), and OCF0 are set (one), the Timer/Counter0 Compare MatchInterrupt is executed.

• Bit 0 – TOV0: Timer/Counter0 Overflow FlagThe bit TOV0 is set (one) when an overflow occurs in Timer/Counter0. TOV0 is cleared by hardware when execut-ing the corresponding interrupt handling vector. Alternatively, TOV0 is cleared by writing a logic one to the flag.When the SREG I-bit, TOIE0 (Timer/Counter0 Overflow Interrupt Enable), and TOV0 are set (one), theTimer/Counter0 Overflow interrupt is executed. In phase correct PWM mode, this bit is set when Timer/Counter0changes counting direction at $00.

Bit 7 6 5 4 3 2 1 0

OCF2 TOV2 ICF1 OCF1A OCF1B TOV1 OCF0 TOV0 TIFR




16. Timer/Counter0 and Timer/Counter1 Prescalers

16.1 OverviewTimer/Counter1 and Timer/Counter0 share the same prescaler module, but the Timer/Counters can have differentprescaler settings. The description below applies to both Timer/Counter1 and Timer/Counter0.

16.2 Internal Clock SourceThe Timer/Counter can be clocked directly by the system clock (by setting the CSn2:0 = 1). This provides the fast-est operation, with a maximum Timer/Counter clock frequency equal to system clock frequency (fCLK_I/O).Alternatively, one of four taps from the prescaler can be used as a clock source. The prescaled clock has a fre-quency of either fCLK_I/O/8, fCLK_I/O/64, fCLK_I/O/256, or fCLK_I/O/1024.

16.3 Prescaler ResetThe prescaler is free running, that is, operates independently of the clock select logic of the Timer/Counter, and it isshared by Timer/Counter1 and Timer/Counter0. Since the prescaler is not affected by the Timer/Counter’s clockselect, the state of the prescaler will have implications for situations where a prescaled clock is used. One exampleof prescaling artifacts occurs when the timer is enabled and clocked by the prescaler (6 > CSn2:0 > 1). The numberof system clock cycles from when the timer is enabled to the first count occurs can be from 1 to N+1 system clockcycles, where N equals the prescaler divisor (8, 64, 256, or 1024).

It is possible to use the Prescaler Reset for synchronizing the Timer/Counter to program execution. However, caremust be taken if the other Timer/Counter that shares the same prescaler also uses prescaling. A prescaler resetwill affect the prescaler period for all Timer/Counters it is connected to.

16.4 External Clock SourceAn external clock source applied to the T1/T0 pin can be used as Timer/Counter clock (clkT1/clkT0). The T1/T0 pinis sampled once every system clock cycle by the pin synchronization logic. The synchronized (sampled) signal isthen passed through the edge detector. Figure 15-1 shows a functional equivalent block diagram of the T1/T0 syn-chronization and edge detector logic. The registers are clocked at the positive edge of the internal system clock(clkI/O). The latch is transparent in the high period of the internal system clock.

The edge detector generates one clkT1/clkT0 pulse for each positive (CSn2:0 = 7) or negative (CSn2:0 = 6) edge itdetects.

Figure 16-1. T1/T0 Pin Sampling

The synchronization and edge detector logic introduces a delay of 2.5 to 3.5 system clock cycles from an edge hasbeen applied to the T1/T0 pin to the counter is updated.

Enabling and disabling of the clock input must be done when T1/T0 has been stable for at least one system clockcycle, otherwise it is a risk that a false Timer/Counter clock pulse is generated.

Tn_sync(To ClockSelect Logic)

Edge DetectorSynchronization

D QD Q

LE

D QTn

clkI/O


Each half period of the external clock applied must be longer than one system clock cycle to ensure correct sam-pling. The external clock must be guaranteed to have less than half the system clock frequency (fExtClk < fclk_I/O/2)given a 50/50% duty cycle. Since the edge detector uses sampling, the maximum frequency of an external clock itcan detect is half the sampling frequency (Nyquist sampling theorem). However, due to variation of the systemclock frequency and duty cycle caused by Oscillator source (crystal, resonator, and capacitors) tolerances, it is rec-ommended that maximum frequency of an external clock source is less than fclk_I/O/2.5.

An external clock source can not be prescaled.

Figure 16-2. Prescaler for Timer/Counter0 and Timer/Counter1(1)

Note: 1. The synchronization logic on the input pins (T1/T0) is shown in Figure 15-1.

PSR10

Clear

clkT1 clkT0

T1

T0

clkI/O

Synchronization

Synchronization



16.5.1 SFIOR – Special Function IO Register

• Bit 0 – PSR10: Prescaler Reset Timer/Counter1 and Timer/Counter0When this bit is written to one, the Timer/Counter1 and Timer/Counter0 prescaler will be reset. The bit will becleared by hardware after the operation is performed. Writing a zero to this bit will have no effect. Note thatTimer/Counter1 and Timer/Counter0 share the same prescaler and a reset of this prescaler will affect both timers.This bit will always be read as zero.

Bit 7 6 5 4 3 2 1 0





17. 16-bit Timer/Counter1

17.1 Features• True 16-bit Design (that is, allows 16-bit PWM)• Two Independent Output Compare Units• Double Buffered Output Compare Registers• One Input Capture Unit• Input Capture Noise Canceler• Clear Timer on Compare Match (Auto Reload)• Glitch-free, Phase Correct Pulse Width Modulator (PWM)• Variable PWM Period• Frequency Generator• External Event Counter• Four Independent Interrupt Sources (TOV1, OCF1A, OCF1B, and ICF1)

17.2 OverviewThe 16-bit Timer/Counter unit allows accurate program execution timing (event management), wave generation,and signal timing measurement. Most register and bit references in this section are written in general form. A lowercase "n" replaces the Timer/Counter number, and a lower case "x" replaces the output compare unit. However,when using the register or bit defines in a program, the precise form must be used, that is, TCNT1 for accessingTimer/Counter1 counter value and so on.

A simplified block diagram of the 16-bit Timer/Counter is shown in Figure 16-1. For the actual placement of I/Opins, refer to Figure 1-1 on page 2. CPU accessible I/O Registers, including I/O bits and I/O pins, are shown inbold. The device-specific I/O Register and bit locations are listed in the “Register Description” on page 112.


Figure 17-1. 16-bit Timer/Counter Block Diagram(1)

Note: 1. Refer to Figure 1-1 on page 2, Table 12-6 on page 59, and Table 12-12 on page 64 for Timer/Counter1 pin place-ment and description.

17.2.1 RegistersThe Timer/Counter (TCNT1), Output Compare Registers (OCR1A/B), and Input Capture Register (ICR1) are all16-bit registers. Special procedures must be followed when accessing the 16-bit registers. These procedures aredescribed in the section “Accessing 16-bit Registers” on page 94. The Timer/Counter Control Registers(TCCR1A/B) are 8-bit registers and have no CPU access restrictions. Interrupt requests (abbreviated to Int.Req. inthe figure) signals are all visible in the Timer Interrupt Flag Register (TIFR). All interrupts are individually maskedwith the Timer Interrupt Mask Register (TIMSK). TIFR and TIMSK are not shown in the figure since these registersare shared by other timer units.

The Timer/Counter can be clocked internally, via the prescaler, or by an external clock source on the T1 pin. TheClock Select logic block controls which clock source and edge the Timer/Counter uses to increment (or decrement)its value. The Timer/Counter is inactive when no clock source is selected. The output from the clock select logic isreferred to as the timer clock (clkT1).

The double buffered Output Compare Registers (OCR1A/B) are compared with the Timer/Counter value at all time.The result of the compare can be used by the Waveform Generator to generate a PWM or variable frequency out-put on the Output Compare pin (OC1A/B). See “Output Compare Units” on page 100. The compare match eventwill also set the Compare Match Flag (OCF1A/B) which can be used to generate an output compare interruptrequest.

Clock Select

Timer/Counter

DAT

AB

US

OCRnA

OCRnB

ICRn

=

=

TCNTn

WaveformGeneration

WaveformGeneration

OCnA

OCnB

NoiseCanceler

ICPn

=

FixedTOP

Values

EdgeDetector

Control Logic

= 0

TOP BOTTOM

Count

Clear

Direction

TOVn(Int.Req.)

OCnA(Int.Req.)

OCnB(Int.Req.)

ICFn (Int.Req.)

TCCRnA TCCRnB

( From AnalogComparator Ouput )

TnEdge

Detector

( From Prescaler )

clkTn


The Input Capture Register can capture the Timer/Counter value at a given external (edge triggered) event oneither the Input Capture Pin (ICP1) or on the Analog Comparator pins (See “Analog Comparator” on page 205.)The Input Capture unit includes a digital filtering unit (Noise Canceler) for reducing the chance of capturing noisespikes.

The TOP value, or maximum Timer/Counter value, can in some modes of operation be defined by either theOCR1A Register, the ICR1 Register, or by a set of fixed values. When using OCR1A as TOP value in a PWMmode, the OCR1A Register can not be used for generating a PWM output. However, the TOP value will in thiscase be double buffered allowing the TOP value to be changed in run time. If a fixed TOP value is required, theICR1 Register can be used as an alternative, freeing the OCR1A to be used as PWM output.

17.2.2 DefinitionsThe following definitions are used extensively throughout the document:

17.2.3 CompatibilityThe 16-bit Timer/Counter has been updated and improved from previous versions of the 16-bit AVR Timer/Coun-ter. This 16-bit Timer/Counter is fully compatible with the earlier version regarding:

• All 16-bit Timer/Counter related I/O Register address locations, including Timer Interrupt Registers.

• Bit locations inside all 16-bit Timer/Counter Registers, including Timer Interrupt Registers.

• Interrupt Vectors.

The following control bits have changed name, but have same functionality and register location:

• PWM10 is changed to WGM10.

• PWM11 is changed to WGM11.

• CTC1 is changed to WGM12.

The following bits are added to the 16-bit Timer/Counter Control Registers:

• FOC1A and FOC1B are added to TCCR1A.

• WGM13 is added to TCCR1B.

The 16-bit Timer/Counter has improvements that will affect the compatibility in some special cases.

17.3 Accessing 16-bit RegistersThe TCNT1, OCR1A/B, and ICR1 are 16-bit registers that can be accessed by the AVR CPU via the 8-bit data bus.The 16-bit register must be byte accessed using two read or write operations. Each 16-bit timer has a single 8-bitregister for temporary storing of the high byte of the 16-bit access. The same temporary register is shared betweenall 16-bit registers within each 16-bit timer. Accessing the low byte triggers the 16-bit read or write operation. Whenthe low byte of a 16-bit register is written by the CPU, the high byte stored in the temporary register, and the lowbyte written are both copied into the 16-bit register in the same clock cycle. When the low byte of a 16-bit register isread by the CPU, the high byte of the 16-bit register is copied into the temporary register in the same clock cycle asthe low byte is read.


BOTTOM The counter reaches the BOTTOM when it becomes 0x0000.

MAX The counter reaches its MAXimum when it becomes 0xFFFF (decimal 65535).

TOP

The counter reaches the TOP when it becomes equal to the highest value in the count sequence. The TOP value can be assigned to be one of the fixed values: 0x00FF, 0x01FF, or 0x03FF, or to the value stored in the OCR1A or ICR1 Register. The assign-ment is dependent of the mode of operation.


Not all 16-bit accesses uses the temporary register for the high byte. Reading the OCR1A/B 16-bit registers doesnot involve using the temporary register.

To do a 16-bit write, the high byte must be written before the low byte. For a 16-bit read, the low byte must be readbefore the high byte.

The following code examples show how to access the 16-bit Timer Registers assuming that no interrupts updatesthe temporary register. The same principle can be used directly for accessing the OCR1A/B and ICR1 Registers.Note that when using “C”, the compiler handles the 16-bit access.

Note: 1. See “About Code Examples” on page 6.

The assembly code example returns the TCNT1 value in the r17:r16 register pair.

It is important to notice that accessing 16-bit registers are atomic operations. If an interrupt occurs between the twoinstructions accessing the 16-bit register, and the interrupt code updates the temporary register by accessing thesame or any other of the 16-bit Timer Registers, then the result of the access outside the interrupt will be corrupted.Therefore, when both the main code and the interrupt code update the temporary register, the main code must dis-able the interrupts during the 16-bit access.


:.

; Set TCNT1 to 0x01FF

ldi r17,0x01

ldi r16,0xFF

out TCNT1H,r17

out TCNT1L,r16

; Read TCNT1 into r17:r16

in r16,TCNT1L

in r17,TCNT1H

:.

C Code Example(1)

unsigned int i;

:.

/* Set TCNT1 to 0x01FF */

TCNT1 = 0x1FF;

/* Read TCNT1 into i */

i = TCNT1;

:.


The following code examples show how to do an atomic read of the TCNT1 Register contents. Reading any of theOCR1A/B or ICR1 Registers can be done by using the same principle.


The assembly code example returns the TCNT1 value in the r17:r16 register pair.


TIM16_ReadTCNT1:

; Save global interrupt flag

in r18,SREG

; Disable interrupts

cli

; Read TCNT1 into r17:r16

in r16,TCNT1L

in r17,TCNT1H

; Restore global interrupt flag

out SREG,r18

ret

C Code Example(1)

unsigned int TIM16_ReadTCNT1( void )

{

unsigned char sreg;

unsigned int i;

/* Save global interrupt flag */

sreg = SREG;

/* Disable interrupts */

_CLI();

/* Read TCNT1 into i */

i = TCNT1;

/* Restore global interrupt flag */

SREG = sreg;

return i;

}


The following code examples show how to do an atomic write of the TCNT1 Register contents. Writing any of theOCR1A/B or ICR1 Registers can be done by using the same principle.


The assembly code example requires that the r17:r16 register pair contains the value to be written to TCNT1.

17.3.1 Reusing the Temporary High Byte RegisterIf writing to more than one 16-bit register where the high byte is the same for all registers written, then the high byteonly needs to be written once. However, note that the same rule of atomic operation described previously alsoapplies in this case.

17.4 Timer/Counter Clock SourcesThe Timer/Counter can be clocked by an internal or an external clock source. The clock source is selected by theClock Select logic which is controlled by the Clock Select (CS12:0) bits located in the Timer/Counter Control Reg-ister B (TCCR1B). For details on clock sources and prescaler, see “Timer/Counter0 and Timer/Counter1Prescalers” on page 88.

17.5 Counter UnitThe main part of the 16-bit Timer/Counter is the programmable 16-bit bi-directional counter unit. Figure 16-2 showsa block diagram of the counter and its surroundings.


TIM16_WriteTCNT1:

; Save global interrupt flag

in r18,SREG

; Disable interrupts

cli

; Set TCNT1 to r17:r16

out TCNT1H,r17

out TCNT1L,r16

; Restore global interrupt flag

out SREG,r18

ret

C Code Example(1)

void TIM16_WriteTCNT1 ( unsigned int i )

{

unsigned char sreg;

unsigned int i;

/* Save global interrupt flag */

sreg = SREG;

/* Disable interrupts */

_CLI();

/* Set TCNT1 to i */

TCNT1 = i;

/* Restore global interrupt flag */

SREG = sreg;

}




Count Increment or decrement TCNT1 by 1.

Direction Select between increment and decrement.

Clear Clear TCNT1 (set all bits to zero).

clkT1 Timer/Counter clock.

TOP Signalize that TCNT1 has reached maximum value.

BOTTOM Signalize that TCNT1 has reached minimum value (zero).

The 16-bit counter is mapped into two 8-bit I/O memory locations: Counter High (TCNT1H) containing the uppereight bits of the counter, and Counter Low (TCNT1L) containing the lower 8 bits. The TCNT1H Register can onlybe indirectly accessed by the CPU. When the CPU does an access to the TCNT1H I/O location, the CPU accessesthe high byte temporary register (TEMP). The temporary register is updated with the TCNT1H value when theTCNT1L is read, and TCNT1H is updated with the temporary register value when TCNT1L is written. This allowsthe CPU to read or write the entire 16-bit counter value within one clock cycle via the 8-bit data bus. It is importantto notice that there are special cases of writing to the TCNT1 Register when the counter is counting that will giveunpredictable results. The special cases are described in the sections where they are of importance.

Depending on the mode of operation used, the counter is cleared, incremented, or decremented at each timerclock (clkT1). The clkT1 can be generated from an external or internal clock source, selected by the Clock Select bits(CS12:0). When no clock source is selected (CS12:0 = 0) the timer is stopped. However, the TCNT1 value can beaccessed by the CPU, independent of whether clkT1 is present or not. A CPU write overrides (has priority over) allcounter clear or count operations.

The counting sequence is determined by the setting of the Waveform Generation Mode bits (WGM13:0) located inthe Timer/Counter Control Registers A and B (TCCR1A and TCCR1B). There are close connections between howthe counter behaves (counts) and how waveforms are generated on the Output Compare outputs OC1x. For moredetails about advanced counting sequences and waveform generation, see “Modes of Operation” on page 103.


17.6 Input Capture UnitThe Timer/Counter incorporates an Input Capture unit that can capture external events and give them a time-stamp indicating time of occurrence. The external signal indicating an event, or multiple events, can be applied viathe ICP1 pin or alternatively, via the Analog Comparator unit. The time-stamps can then be used to calculate fre-quency, duty-cycle, and other features of the signal applied. Alternatively the time-stamps can be used for creatinga log of the events.

TEMP (8-bit)

DATA BUS (8-bit)

TCNTn (16-bit Counter)

TCNTnH (8-bit) TCNTnL (8-bit)Control Logic

Count

Clear

Direction

TOVn(Int.Req.)

Clock Select

TOP BOTTOM

TnEdge

Detector

( From Prescaler )

clkTn


The Input Capture unit is illustrated by the block diagram shown in Figure 16-3. The elements of the block diagramthat are not directly a part of the Input Capture unit are gray shaded. The small “n” in register and bit names indi-cates the Timer/Counter number.

Figure 17-3. Input Capture Unit Block Diagram

When a change of the logic level (an event) occurs on the Input Capture pin (ICP1), alternatively on the AnalogComparator output (ACO), and this change confirms to the setting of the edge detector, a capture will be triggered.When a capture is triggered, the 16-bit value of the counter (TCNT1) is written to the Input Capture Register(ICR1). The Input Capture Flag (ICF1) is set at the same system clock as the TCNT1 value is copied into ICR1Register. If enabled (TICIE1 = 1), the Input Capture Flag generates an Input Capture interrupt. The ICF1 Flag isautomatically cleared when the interrupt is executed. Alternatively the ICF1 Flag can be cleared by software bywriting a logical one to its I/O bit location.

Reading the 16-bit value in the Input Capture Register (ICR1) is done by first reading the low byte (ICR1L) andthen the high byte (ICR1H). When the low byte is read the high byte is copied into the high byte temporary register(TEMP). When the CPU reads the ICR1H I/O location it will access the TEMP Register.

The ICR1 Register can only be written when using a Waveform Generation mode that utilizes the ICR1 Register fordefining the counter’s TOP value. In these cases the Waveform Generation mode (WGM1[3:0]) bits must be setbefore the TOP value can be written to the ICR1 Register. When writing the ICR1 Register the high byte must bewritten to the ICR1H I/O location before the low byte is written to ICR1L.

For more information on how to access the 16-bit registers refer to “Accessing 16-bit Registers” on page 94.

17.6.1 Input Capture Pin SourceThe main trigger source for the Input Capture unit is the Input Capture pin (ICP1). Timer/Counter1 can alternativelyuse the Analog Comparator output as trigger source for the Input Capture unit. The Analog Comparator is selectedas trigger source by setting the Analog Comparator Input Capture (ACIC) bit in the Analog Comparator Control andStatus Register (ACSR). Be aware that changing trigger source can trigger a capture. The Input Capture Flag musttherefore be cleared after the change.

ICFn (Int.Req.)

AnalogComparator

WRITE ICRn (16-bit Register)

ICRnH (8-bit)

NoiseCanceler

ICPn

EdgeDetector

TEMP (8-bit)

DATA BUS (8-bit)

ICRnL (8-bit)


TCNTnH (8-bit) TCNTnL (8-bit)

ACIC* ICNC ICESACO*


Both the Input Capture pin (ICP1) and the Analog Comparator output (ACO) inputs are sampled using the sametechnique as for the T1 pin (Figure 15-1 on page 89). The edge detector is also identical. However, when the noisecanceler is enabled, additional logic is inserted before the edge detector, which increases the delay by four systemclock cycles. Note that the input of the noise canceler and edge detector is always enabled unless the Timer/Coun-ter is set in a waveform generation mode that uses ICR1 to define TOP.

An Input Capture can be triggered by software by controlling the port of the ICP1 pin.

17.6.2 Noise CancelerThe noise canceler improves noise immunity by using a simple digital filtering scheme. The noise canceler input ismonitored over four samples, and all four must be equal for changing the output that in turn is used by the edgedetector.

The noise canceler is enabled by setting the Input Capture Noise Canceler (ICNC1) bit in Timer/Counter ControlRegister B (TCCR1B). When enabled the noise canceler introduces additional four system clock cycles of delayfrom a change applied to the input, to the update of the ICR1 Register. The noise canceler uses the system clockand is therefore not affected by the prescaler.

17.6.3 Using the Input Capture UnitThe main challenge when using the Input Capture unit is to assign enough processor capacity for handling theincoming events. The time between two events is critical. If the processor has not read the captured value in theICR1 Register before the next event occurs, the ICR1 will be overwritten with a new value. In this case the result ofthe capture will be incorrect.

When using the Input Capture interrupt, the ICR1 Register should be read as early in the interrupt handler routineas possible. Even though the Input Capture interrupt has relatively high priority, the maximum interrupt responsetime is dependent on the maximum number of clock cycles it takes to handle any of the other interrupt requests.

Using the Input Capture unit in any mode of operation when the TOP value (resolution) is actively changed duringoperation, is not recommended.

Measurement of an external signal’s duty cycle requires that the trigger edge is changed after each capture.Changing the edge sensing must be done as early as possible after the ICR1 Register has been read. After achange of the edge, the Input Capture Flag (ICF1) must be cleared by software (writing a logical one to the I/O bitlocation). For measuring frequency only, the clearing of the ICF1 Flag is not required (if an interrupt handler isused).

17.6.4 Output Compare UnitsThe 16-bit comparator continuously compares TCNT1 with the Output Compare Register (OCR1x). If TCNT equalsOCR1x the comparator signals a match. A match will set the Output Compare Flag (OCF1x) at the next timer clockcycle. If enabled (OCIE1x = 1), the Output Compare Flag generates an output compare interrupt. The OCF1x Flagis automatically cleared when the interrupt is executed. Alternatively the OCF1x Flag can be cleared by softwareby writing a logical one to its I/O bit location. The Waveform Generator uses the match signal to generate an outputaccording to operating mode set by the Waveform Generation mode (WGM13:0) bits and Compare Output mode(COM1x1:0) bits. The TOP and BOTTOM signals are used by the Waveform Generator for handling the specialcases of the extreme values in some modes of operation (See “Modes of Operation” on page 103.)

A special feature of output compare unit A allows it to define the Timer/Counter TOP value (that is, counter resolu-tion). In addition to the counter resolution, the TOP value defines the period time for waveforms generated by theWaveform Generator.

Figure 16-4 shows a block diagram of the output compare unit. The small “n” in the register and bit names indicatesthe device number (n = 1 for Timer/Counter1), and the “x” indicates output compare unit (A/B). The elements of theblock diagram that are not directly a part of the output compare unit are gray shaded.



The OCR1x Register is double buffered when using any of the twelve Pulse Width Modulation (PWM) modes. Forthe normal and Clear Timer on Compare (CTC) modes of operation, the double buffering is disabled. The doublebuffering synchronizes the update of the OCR1x Compare Register to either TOP or BOTTOM of the countingsequence. The synchronization prevents the occurrence of odd-length, non-symmetrical PWM pulses, therebymaking the output glitch-free.

The OCR1x Register access may seem complex, but this is not case. When the double buffering is enabled, theCPU has access to the OCR1x Buffer Register, and if double buffering is disabled the CPU will access the OCR1xdirectly. The content of the OCR1x (Buffer or Compare) Register is only changed by a write operation (theTimer/Counter does not update this register automatically as the TCNT1 and ICR1 Register). Therefore OCR1x isnot read via the high byte temporary register (TEMP). However, it is a good practice to read the low byte first aswhen accessing other 16-bit registers. Writing the OCR1x Registers must be done via the TEMP Register since thecompare of all 16 bits is done continuously. The high byte (OCR1xH) has to be written first. When the high byte I/Olocation is written by the CPU, the TEMP Register will be updated by the value written. Then when the low byte(OCR1xL) is written to the lower eight bits, the high byte will be copied into the upper 8-bits of either the OCR1xbuffer or OCR1x Compare Register in the same system clock cycle.

For more information of how to access the 16-bit registers refer to “Accessing 16-bit Registers” on page 94.

17.6.5 Force Output CompareIn non-PWM Waveform Generation modes, the match output of the comparator can be forced by writing a one tothe Force Output Compare (FOC1x) bit. Forcing compare match will not set the OCF1x Flag or reload/clear thetimer, but the OC1x pin will be updated as if a real compare match had occurred (the COM1x1:0 bits settingsdefine whether the OC1x pin is set, cleared or toggled).

17.6.6 Compare Match Blocking by TCNT1 WriteAll CPU writes to the TCNT1 Register will block any compare match that occurs in the next timer clock cycle, evenwhen the timer is stopped. This feature allows OCR1x to be initialized to the same value as TCNT1 without trigger-ing an interrupt when the Timer/Counter clock is enabled.

OCFnx (Int.Req.)


OCRnx Buffer (16-bit Register)

OCRnxH Buf. (8-bit)

OCnx

TEMP (8-bit)

DATA BUS (8-bit)

OCRnxL Buf. (8-bit)


TCNTnH (8-bit) TCNTnL (8-bit)

COMnx1:0WGMn3:0

OCRnx (16-bit Register)

OCRnxH (8-bit) OCRnxL (8-bit)

Waveform GeneratorTOP

BOTTOM


17.6.7 Using the Output Compare UnitSince writing TCNT1 in any mode of operation will block all compare matches for one timer clock cycle, there arerisks involved when changing TCNT1 when using any of the output compare units, independent of whether theTimer/Counter is running or not. If the value written to TCNT1 equals the OCR1x value, the compare match will bemissed, resulting in incorrect waveform generation. Do not write the TCNT1 equal to TOP in PWM modes with vari-able TOP values. The compare match for the TOP will be ignored and the counter will continue to 0xFFFF.Similarly, do not write the TCNT1 value equal to BOTTOM when the counter is down counting.

The setup of the OC1x should be performed before setting the Data Direction Register for the port pin to output.The easiest way of setting the OC1x value is to use the force output compare (FOC1x) strobe bits in Normal mode.The OC1x Register keeps its value even when changing between waveform generation modes.

Be aware that the COM1x1:0 bits are not double buffered together with the compare value. Changing theCOM1x1:0 bits will take effect immediately.

17.7 Compare Match Output UnitThe Compare Output mode (COM1x1:0) bits have two functions. The Waveform Generator uses the COM1x1:0bits for defining the Output Compare (OC1x) state at the next compare match. Secondly the COM1x1:0 bits controlthe OC1x pin output source. Figure 16-5 shows a simplified schematic of the logic affected by the COM1x1:0 bitsetting. The I/O Registers, I/O bits, and I/O pins in the figure are shown in bold. Only the parts of the general I/OPort Control Registers (DDR and PORT) that are affected by the COM1x1:0 bits are shown. When referring to theOC1x state, the reference is for the internal OC1x Register, not the OC1x pin. If a System Reset occur, the OC1xRegister is reset to “0”.


The general I/O port function is overridden by the Output Compare (OC1x) from the Waveform Generator if eitherof the COM1x1:0 bits are set. However, the OC1x pin direction (input or output) is still controlled by the Data Direc-tion Register (DDR) for the port pin. The Data Direction Register bit for the OC1x pin (DDR_OC1x) must be set asoutput before the OC1x value is visible on the pin. The port override function is generally independent of the Wave-form Generation mode, but there are some exceptions. Refer to Table 16-2, Table 16-3 and Table 16-4 for details.

PORT

DDR

D Q

D Q

OCnxPinOCnx


COMnx1

COMnx0

0

1

DAT

AB

US

FOCnx

clkI/O


The design of the output compare pin logic allows initialization of the OC1x state before the output is enabled. Notethat some COM1x1:0 bit settings are reserved for certain modes of operation. See “Register Description” on page112.

The COM1x1:0 bits have no effect on the Input Capture unit.

17.7.1 Compare Output Mode and Waveform GenerationThe Waveform Generator uses the COM1x1:0 bits differently in normal, CTC, and PWM modes. For all modes,setting the COM1x1:0 = 0 tells the Waveform Generator that no action on the OC1x Register is to be performed onthe next compare match. For compare output actions in the non-PWM modes refer to Table 16-2 on page 112. Forfast PWM mode refer to Table 16-3 on page 113, and for phase correct and phase and frequency correct PWMrefer to Table 16-4 on page 113.

A change of the COM1x1:0 bits state will have effect at the first compare match after the bits are written. For non-PWM modes, the action can be forced to have immediate effect by using the FOC1x strobe bits.

17.8 Modes of OperationThe mode of operation, that is, the behavior of the Timer/Counter and the output compare pins, is defined by thecombination of the Waveform Generation mode (WGM13:0) and Compare Output mode (COM1x1:0) bits. TheCompare Output mode bits do not affect the counting sequence, while the Waveform Generation mode bits do.The COM1x1:0 bits control whether the PWM output generated should be inverted or not (inverted or non-invertedPWM). For non-PWM modes the COM1x1:0 bits control whether the output should be set, cleared or toggle at acompare match (See “Compare Match Output Unit” on page 102.)

For detailed timing information refer to “Timer/Counter Timing Diagrams” on page 110.

17.8.1 Normal ModeThe simplest mode of operation is the Normal mode (WGM13:0 = 0). In this mode the counting direction is alwaysup (incrementing), and no counter clear is performed. The counter simply overruns when it passes its maximum16-bit value (MAX = 0xFFFF) and then restarts from the BOTTOM (0x0000). In normal operation the Timer/Coun-ter Overflow Flag (TOV1) will be set in the same timer clock cycle as the TCNT1 becomes zero. The TOV1 Flag inthis case behaves like a 17th bit, except that it is only set, not cleared. However, combined with the timer overflowinterrupt that automatically clears the TOV1 Flag, the timer resolution can be increased by software. There are nospecial cases to consider in the Normal mode, a new counter value can be written anytime.

The Input Capture unit is easy to use in Normal mode. However, observe that the maximum interval between theexternal events must not exceed the resolution of the counter. If the interval between events are too long, the timeroverflow interrupt or the prescaler must be used to extend the resolution for the capture unit.

The output compare units can be used to generate interrupts at some given time. Using the output compare to gen-erate waveforms in Normal mode is not recommended, since this will occupy too much of the CPU time.

17.8.2 Clear Timer on Compare Match (CTC) ModeIn Clear Timer on Compare or CTC mode (WGM13:0 = 4 or 12), the OCR1A or ICR1 Register are used to manipu-late the counter resolution. In CTC mode the counter is cleared to zero when the counter value (TCNT1) matcheseither the OCR1A (WGM13:0 = 4) or the ICR1 (WGM13:0 = 12). The OCR1A or ICR1 define the top value for thecounter, hence also its resolution. This mode allows greater control of the compare match output frequency. It alsosimplifies the operation of counting external events.

The timing diagram for the CTC mode is shown in Figure 16-6. The counter value (TCNT1) increases until a com-pare match occurs with either OCR1A or ICR1, and then counter (TCNT1) is cleared.



An interrupt can be generated at each time the counter value reaches the TOP value by either using the OCF1A orICF1 Flag according to the register used to define the TOP value. If the interrupt is enabled, the interrupt handlerroutine can be used for updating the TOP value. However, changing the TOP to a value close to BOTTOM whenthe counter is running with none or a low prescaler value must be done with care since the CTC mode does nothave the double buffering feature. If the new value written to OCR1A or ICR1 is lower than the current value ofTCNT1, the counter will miss the compare match. The counter will then have to count to its maximum value(0xFFFF) and wrap around starting at 0x0000 before the compare match can occur. In many cases this feature isnot desirable. An alternative will then be to use the fast PWM mode using OCR1A for defining TOP (WGM13:0 =15) since the OCR1A then will be double buffered.

For generating a waveform output in CTC mode, the OC1A output can be set to toggle its logical level on eachcompare match by setting the compare output mode bits to toggle mode (COM1A1:0 = 1). The OC1A value will notbe visible on the port pin unless the data direction for the pin is set to output (DDR_OC1A = 1). The waveform gen-erated will have a maximum frequency of fOC1A = fclk_I/O/2 when OCR1A is set to zero (0x0000). The waveformfrequency is defined by the following equation:

The N variable represents the prescaler factor (1, 8, 64, 256, or 1024).


17.8.3 Fast PWM ModeThe fast Pulse Width Modulation or fast PWM mode (WGM13:0 = 5, 6, 7, 14, or 15) provides a high frequencyPWM waveform generation option. The fast PWM differs from the other PWM options by its single-slope operation.The counter counts from BOTTOM to TOP then restarts from BOTTOM. In non-inverting Compare Output mode,the Output Compare (OC1x) is cleared on the compare match between TCNT1 and OCR1x, and set at BOTTOM.In inverting Compare Output mode output is set on compare match and cleared at BOTTOM. Due to the single-slope operation, the operating frequency of the fast PWM mode can be twice as high as the phase correct andphase and frequency correct PWM modes that use dual-slope operation. This high frequency makes the fast PWMmode well suited for power regulation, rectification, and DAC applications. High frequency allows physically smallsized external components (coils, capacitors), hence reduces total system cost.

TCNTn

OCnA(Toggle)

OCnA Interrupt Flag Setor ICFn Interrupt Flag Set(Interrupt on TOP)

1 4Period 2 3

(COMnA1:0 = 1)

fOCnA

fclk_I/O

2 N 1 OCRnA+ ---------------------------------------------------=


The PWM resolution for fast PWM can be fixed to 8-bit, 9-bit, or 10-bit, or defined by either ICR1 or OCR1A. Theminimum resolution allowed is 2-bit (ICR1 or OCR1A set to 0x0003), and the maximum resolution is 16-bit (ICR1 orOCR1A set to MAX). The PWM resolution in bits can be calculated by using the following equation:

In fast PWM mode the counter is incremented until the counter value matches either one of the fixed values0x00FF, 0x01FF, or 0x03FF (WGM13:0 = 5, 6, or 7), the value in ICR1 (WGM13:0 = 14), or the value in OCR1A(WGM13:0 = 15). The counter is then cleared at the following timer clock cycle. The timing diagram for the fastPWM mode is shown in Figure 16-7. The figure shows fast PWM mode when OCR1A or ICR1 is used to defineTOP. The TCNT1 value is in the timing diagram shown as a histogram for illustrating the single-slope operation.The diagram includes non-inverted and inverted PWM outputs. The small horizontal line marks on the TCNT1slopes represent compare matches between OCR1x and TCNT1. The OC1x Interrupt Flag will be set when a com-pare match occurs.


The Timer/Counter Overflow Flag (TOV1) is set each time the counter reaches TOP. In addition the OC1A or ICF1Flag is set at the same timer clock cycle as TOV1 is set when either OCR1A or ICR1 is used for defining the TOPvalue. If one of the interrupts are enabled, the interrupt handler routine can be used for updating the TOP and com-pare values.

When changing the TOP value the program must ensure that the new TOP value is higher or equal to the value ofall of the Compare Registers. If the TOP value is lower than any of the Compare Registers, a compare match willnever occur between the TCNT1 and the OCR1x. Note that when using fixed TOP values the unused bits aremasked to zero when any of the OCR1x Registers are written.

The procedure for updating ICR1 differs from updating OCR1A when used for defining the TOP value. The ICR1Register is not double buffered. This means that if ICR1 is changed to a low value when the counter is running withnone or a low prescaler value, there is a risk that the new ICR1 value written is lower than the current value ofTCNT1. The result will then be that the counter will miss the compare match at the TOP value. The counter willthen have to count to the MAX value (0xFFFF) and wrap around starting at 0x0000 before the compare match canoccur. The OCR1A Register however, is double buffered. This feature allows the OCR1A I/O location to be writtenanytime. When the OCR1A I/O location is written the value written will be put into the OCR1A Buffer Register. TheOCR1A Compare Register will then be updated with the value in the Buffer Register at the next timer clock cycle

RFPWMTOP 1+ log

2 log-----------------------------------=

TCNTn

OCRnx / TOP Update andTOVn Interrupt Flag Set andOCnA Interrupt Flag SetOCnA Interrupt Flag Set(Interrupt on TOP)

1 7Period 2 3 4 5 6 8

OCnx

OCnx

(COMnx1:0 = 2)

(COMnx1:0 = 3)


the TCNT1 matches TOP. The update is done at the same timer clock cycle as the TCNT1 is cleared and theTOV1 Flag is set.

Using the ICR1 Register for defining TOP works well when using fixed TOP values. By using ICR1, the OCR1ARegister is free to be used for generating a PWM output on OC1A. However, if the base PWM frequency is activelychanged (by changing the TOP value), using the OCR1A as TOP is clearly a better choice due to its double bufferfeature.

In fast PWM mode, the compare units allow generation of PWM waveforms on the OC1x pins. Setting theCOM1x1:0 bits to 2 will produce a non-inverted PWM and an inverted PWM output can be generated by setting theCOM1x1:0 to 3 (See Table 16-2 on page 112). The actual OC1x value will only be visible on the port pin if the datadirection for the port pin is set as output (DDR_OC1x). The PWM waveform is generated by seting (or clearing) theOC1x Register at the compare match between OCR1x and TCNT1, and clearing (or setting) the OC1x Register atthe timer clock cycle the counter is cleared (changes from TOP to BOTTOM).


The N variable represents the prescaler divider (1, 8, 64, 256, or 1024).

The extreme values for the OCR1x Register represents special cases when generating a PWM waveform output inthe fast PWM mode. If the OCR1x is set equal to BOTTOM (0x0000) the output will be a narrow spike for eachTOP+1 timer clock cycle. Setting the OCR1x equal to TOP will result in a constant high or low output (dependingon the polarity of the output set by the COM1x1:0 bits.)

A frequency (with 50% duty cycle) waveform output in fast PWM mode can be achieved by setting OC1A to toggleits logical level on each compare match (COM1A1:0 = 1). This applies only if OCR1A is used to define the TOPvalue (WGM13:0 = 15). The waveform generated will have a maximum frequency of fOC1A = fclk_I/O/2 when OCR1Ais set to zero (0x0000). This feature is similar to the OC1A toggle in CTC mode, except the double buffer feature ofthe output compare unit is enabled in the fast PWM mode.

17.8.4 Phase Correct PWM ModeThe phase correct Pulse Width Modulation or phase correct PWM mode (WGM13:0 = 1, 2, 3, 10, or 11) provides ahigh resolution phase correct PWM waveform generation option. The phase correct PWM mode is, like the phaseand frequency correct PWM mode, based on a dual-slope operation. The counter counts repeatedly from BOT-TOM (0x0000) to TOP and then from TOP to BOTTOM. In non-inverting Compare Output mode, the OutputCompare (OC1x) is cleared on the compare match between TCNT1 and OCR1x while upcounting, and set on thecompare match while downcounting. In inverting Output Compare mode, the operation is inverted. The dual-slopeoperation has lower maximum operation frequency than single slope operation. However, due to the symmetricfeature of the dual-slope PWM modes, these modes are preferred for motor control applications.

The PWM resolution for the phase correct PWM mode can be fixed to 8-bit, 9-bit, or 10-bit, or defined by eitherICR1 or OCR1A. The minimum resolution allowed is 2-bit (ICR1 or OCR1A set to 0x0003), and the maximum res-olution is 16-bit (ICR1 or OCR1A set to MAX). The PWM resolution in bits can be calculated by using the followingequation:

In phase correct PWM mode the counter is incremented until the counter value matches either one of the fixed val-ues 0x00FF, 0x01FF, or 0x03FF (WGM13:0 = 1, 2, or 3), the value in ICR1 (WGM13:0 = 10), or the value inOCR1A (WGM13:0 = 11). The counter has then reached the TOP and changes the count direction. The TCNT1value will be equal to TOP for one timer clock cycle. The timing diagram for the phase correct PWM mode is shownon Figure 16-8. The figure shows phase correct PWM mode when OCR1A or ICR1 is used to define TOP. The

fOCnxPWM

fclk_I/O

N 1 TOP+ -----------------------------------=

RPCPWMTOP 1+ log

2 log-----------------------------------=


TCNT1 value is in the timing diagram shown as a histogram for illustrating the dual-slope operation. The diagramincludes non-inverted and inverted PWM outputs. The small horizontal line marks on the TCNT1 slopes representcompare matches between OCR1x and TCNT1. The OC1x Interrupt Flag will be set when a compare matchoccurs.


The Timer/Counter Overflow Flag (TOV1) is set each time the counter reaches BOTTOM. When either OCR1A orICR1 is used for defining the TOP value, the OC1A or ICF1 Flag is set accordingly at the same timer clock cycle asthe OCR1x Registers are updated with the double buffer value (at TOP). The Interrupt Flags can be used to gener-ate an interrupt each time the counter reaches the TOP or BOTTOM value.

When changing the TOP value the program must ensure that the new TOP value is higher or equal to the value ofall of the Compare Registers. If the TOP value is lower than any of the Compare Registers, a compare match willnever occur between the TCNT1 and the OCR1x. Note that when using fixed TOP values, the unused bits aremasked to zero when any of the OCR1x Registers are written. As the third period shown in Figure 16-8 illustrates,changing the TOP actively while the Timer/Counter is running in the phase correct mode can result in an unsym-metrical output. The reason for this can be found in the time of update of the OCR1x Register. Since the OCR1xupdate occurs at TOP, the PWM period starts and ends at TOP. This implies that the length of the falling slope isdetermined by the previous TOP value, while the length of the rising slope is determined by the new TOP value.When these two values differ the two slopes of the period will differ in length. The difference in length gives theunsymmetrical result on the output.

It is recommended to use the phase and frequency correct mode instead of the phase correct mode when chang-ing the TOP value while the Timer/Counter is running. When using a static TOP value there are practically nodifferences between the two modes of operation.

In phase correct PWM mode, the compare units allow generation of PWM waveforms on the OC1x pins. Settingthe COM1x1:0 bits to 2 will produce a non-inverted PWM and an inverted PWM output can be generated by settingthe COM1x1:0 to 3 (See Table 16-2 on page 112). The actual OC1x value will only be visible on the port pin if thedata direction for the port pin is set as output (DDR_OC1x). The PWM waveform is generated by setting (or clear-ing) the OC1x Register at the compare match between OCR1x and TCNT1 when the counter increments, andclearing (or setting) the OC1x Register at compare match between OCR1x and TCNT1 when the counter decre-

OCRnx/TOP Update andOCnA Interrupt Flag Setor ICFn Interrupt Flag Set(Interrupt on TOP)

1 2 3 4

TOVn Interrupt Flag Set(Interrupt on Bottom)

TCNTn

Period

OCnx

OCnx

(COMnx1:0 = 2)

(COMnx1:0 = 3)


ments. The PWM frequency for the output when using phase correct PWM can be calculated by the followingequation:


The extreme values for the OCR1x Register represent special cases when generating a PWM waveform output inthe phase correct PWM mode. If the OCR1x is set equal to BOTTOM the output will be continuously low and if setequal to TOP the output will be continuously high for non-inverted PWM mode. For inverted PWM the output willhave the opposite logic values. If OCR1A is used to define the TOP value (WGM13:0 = 11) and COM1A1:0 = 1, theOC1A output will toggle with a 50% duty cycle.

17.8.5 Phase and Frequency Correct PWM ModeThe phase and frequency correct Pulse Width Modulation, or phase and frequency correct PWM mode (WGM13:0= 8 or 9) provides a high resolution phase and frequency correct PWM waveform generation option. The phaseand frequency correct PWM mode is, like the phase correct PWM mode, based on a dual-slope operation. Thecounter counts repeatedly from BOTTOM (0x0000) to TOP and then from TOP to BOTTOM. In non-inverting Com-pare Output mode, the Output Compare (OC1x) is cleared on the compare match between TCNT1 and OCR1xwhile upcounting, and set on the compare match while downcounting. In inverting Compare Output mode, theoperation is inverted. The dual-slope operation gives a lower maximum operation frequency compared to the sin-gle-slope operation. However, due to the symmetric feature of the dual-slope PWM modes, these modes arepreferred for motor control applications.

The main difference between the phase correct, and the phase and frequency correct PWM mode is the time theOCR1x Register is updated by the OCR1x Buffer Register, (see Figure 16-8 and Figure 16-9).

The PWM resolution for the phase and frequency correct PWM mode can be defined by either ICR1 or OCR1A.The minimum resolution allowed is 2-bit (ICR1 or OCR1A set to 0x0003), and the maximum resolution is 16-bit(ICR1 or OCR1A set to MAX). The PWM resolution in bits can be calculated using the following equation:

In phase and frequency correct PWM mode the counter is incremented until the counter value matches either thevalue in ICR1 (WGM13:0 = 8), or the value in OCR1A (WGM13:0 = 9). The counter has then reached the TOP andchanges the count direction. The TCNT1 value will be equal to TOP for one timer clock cycle. The timing diagramfor the phase correct and frequency correct PWM mode is shown on Figure 16-9. The figure shows phase and fre-quency correct PWM mode when OCR1A or ICR1 is used to define TOP. The TCNT1 value is in the timingdiagram shown as a histogram for illustrating the dual-slope operation. The diagram includes non-inverted andinverted PWM outputs. The small horizontal line marks on the TCNT1 slopes represent compare matches betweenOCR1x and TCNT1. The OC1x Interrupt Flag will be set when a compare match occurs.

fOCnxPCPWM

fclk_I/O

2 N TOP ----------------------------=

RPFCPWMTOP 1+ log

2 log-----------------------------------=


Figure 17-9. Phase and Frequency Correct PWM Mode, Timing Diagram

The Timer/Counter Overflow Flag (TOV1) is set at the same timer clock cycle as the OCR1x Registers are updatedwith the double buffer value (at BOTTOM). When either OCR1A or ICR1 is used for defining the TOP value, theOC1A or ICF1 Flag set when TCNT1 has reached TOP. The Interrupt Flags can then be used to generate an inter-rupt each time the counter reaches the TOP or BOTTOM value.

When changing the TOP value the program must ensure that the new TOP value is higher or equal to the value ofall of the Compare Registers. If the TOP value is lower than any of the Compare Registers, a compare match willnever occur between the TCNT1 and the OCR1x.

As Figure 16-9 shows the output generated is, in contrast to the phase correct mode, symmetrical in all periods.Since the OCR1x Registers are updated at BOTTOM, the length of the rising and the falling slopes will always beequal. This gives symmetrical output pulses and is therefore frequency correct.

Using the ICR1 Register for defining TOP works well when using fixed TOP values. By using ICR1, the OCR1ARegister is free to be used for generating a PWM output on OC1A. However, if the base PWM frequency is activelychanged by changing the TOP value, using the OCR1A as TOP is clearly a better choice due to its double bufferfeature.

In phase and frequency correct PWM mode, the compare units allow generation of PWM waveforms on the OC1xpins. Setting the COM1x1:0 bits to 2 will produce a non-inverted PWM and an inverted PWM output can be gener-ated by setting the COM1x1:0 to 3 (See Table on page 113). The actual OC1x value will only be visible on the portpin if the data direction for the port pin is set as output (DDR_OC1x). The PWM waveform is generated by setting(or clearing) the OC1x Register at the compare match between OCR1x and TCNT1 when the counter increments,and clearing (or setting) the OC1x Register at compare match between OCR1x and TCNT1 when the counter dec-rements. The PWM frequency for the output when using phase and frequency correct PWM can be calculated bythe following equation:


The extreme values for the OCR1x Register represents special cases when generating a PWM waveform output inthe phase correct PWM mode. If the OCR1x is set equal to BOTTOM the output will be continuously low and if setequal to TOP the output will be set to high for non-inverted PWM mode. For inverted PWM the output will have the

OCRnx / TOP UpdateandTOVn Interrupt Flag Set(Interrupt on Bottom)

OCnA Interrupt Flag Setor ICFn Interrupt Flag Set(Interrupt on TOP)

1 2 3 4

TCNTn

Period

OCnx

OCnx

(COMnx1:0 = 2)

(COMnx1:0 = 3)

fOCnxPFCPWM

fclk_I/O

2 N TOP ----------------------------=


opposite logic values. If OCR1A is used to define the TOP value (WGM13:0 = 9) and COM1A1:0 = 1, the OC1Aoutput will toggle with a 50% duty cycle.

17.9 Timer/Counter Timing DiagramsThe Timer/Counter is a synchronous design and the timer clock (clkT1) is therefore shown as a clock enable signalin the following figures. The figures include information on when Interrupt Flags are set, and when the OCR1x Reg-ister is updated with the OCR1x buffer value (only for modes utilizing double buffering). Figure 16-10 shows atiming diagram for the setting of OCF1x.

Figure 17-10. Timer/Counter Timing Diagram, Setting of OCF1x, No Prescaling


Figure 17-11. Timer/Counter Timing Diagram, Setting of OCF1x, with Prescaler (fclk_I/O/8)

Figure 16-12 shows the count sequence close to TOP in various modes. When using phase and frequency correctPWM mode the OCR1x Register is updated at BOTTOM. The timing diagrams will be the same, but TOP shouldbe replaced by BOTTOM, TOP-1 by BOTTOM+1 and so on. The same renaming applies for modes that set theTOV1 Flag at BOTTOM.

clkTn(clkI/O/1)

OCFnx

clkI/O

OCRnx

TCNTn

OCRnx Value

OCRnx - 1 OCRnx OCRnx + 1 OCRnx + 2

OCFnx

OCRnx

TCNTn

OCRnx Value

OCRnx - 1 OCRnx OCRnx + 1 OCRnx + 2

clkI/O

clkTn(clkI/O/8)






17.10.1 TCCR1A – Timer/Counter1 Control Register A

• Bit 7:6 – COM1A1:0: Compare Output Mode for Compare unit A

• Bit 5:4 – COM1B1:0: Compare Output Mode for Compare unit B

TOVn (FPWM)and ICFn (if used

as TOP)

OCRnx(Update at TOP)

TCNTn(CTC and FPWM)

TCNTn(PC and PFC PWM)

TOP - 1 TOP TOP - 1 TOP - 2

Old OCRnx Value New OCRnx Value


clkTn(clkI/O/1)

clkI/O

TOVn (FPWM)and ICFn (if used

as TOP)

OCRnx(Update at TOP)

TCNTn(CTC and FPWM)

TCNTn(PC and PFC PWM)

TOP - 1 TOP TOP - 1 TOP - 2

Old OCRnx Value New OCRnx Value


clkI/O

clkTn(clkI/O/8)

Bit 7 6 5 4 3 2 1 0

COM1A1 COM1A0 COM1B1 COM1B0 FOC1A FOC1B WGM11 WGM10 TCCR1A

Read/Write R/W R/W R/W R/W W W R/W R/W



The COM1A1:0 and COM1B1:0 control the Output Compare pins (OC1A and OC1B respectively) behavior. If oneor both of the COM1A1:0 bits are written to one, the OC1A output overrides the normal port functionality of the I/Opin it is connected to. If one or both of the COM1B1:0 bit are written to one, the OC1B output overrides the normalport functionality of the I/O pin it is connected to. However, note that the Data Direction Register (DDR) bit corre-sponding to the OC1A or OC1B pin must be set in order to enable the output driver.

When the OC1A or OC1B is connected to the pin, the function of the COM1x1:0 bits is dependent of the WGM13:0bits setting. Table 16-2 shows the COM1x1:0 bit functionality when the WGM13:0 bits are set to a normal or a CTCmode (non-PWM).

Table 16-3 shows the COM1x1:0 bit functionality when the WGM13:0 bits are set to the fast PWM mode.

Note: 1. A special case occurs when OCR1A/OCR1B equals TOP and COM1A1/COM1B1 is set. In this case the compare match is ignored, but the set or clear is done at BOTTOM. See “Fast PWM Mode” on page 104. for more details.

Table 17-2. Compare Output Mode, non-PWM

COM1A1/COM1B1 COM1A0/COM1B0 Description

0 0 Normal port operation, OC1A/OC1B disconnected.

0 1 Toggle OC1A/OC1B on compare match

1 0 Clear OC1A/OC1B on compare match (Set output to low level)

1 1 Set OC1A/OC1B on compare match (Set output to high level)

Table 17-3. Compare Output Mode, Fast PWM(1)



0 1 WGM13:0 = 15: Toggle OC1A on Compare Match, OC1B disconnected (normal port operation).For all other WGM13:0 settings, normal port operation, OC1A/OC1B disconnected.

1 0 Clear OC1A/OC1B on compare match, set OC1A/OC1B at BOTTOM,

(non-inverting mode)

1 1 Set OC1A/OC1B on compare match, clear OC1A/OC1B at BOTTOM,(inverting mode)


Table 16-4 shows the COM1x1:0 bit functionality when the WGM13:0 bits are set to the phase correct or the phaseand frequency correct, PWM mode.

Note: 1. A special case occurs when OCR1A/OCR1B equals TOP and COM1A1/COM1B1 is set. See “Phase Correct PWMMode” on page 106. for more details.

• Bit 3 – FOC1A: Force Output Compare for Compare unit A

• Bit 2 – FOC1B: Force Output Compare for Compare unit BThe FOC1A/FOC1B bits are only active when the WGM13:0 bits specifies a non-PWM mode. However, for ensur-ing compatibility with future devices, these bits must be set to zero when TCCR1A is written when operating in aPWM mode. When writing a logical one to the FOC1A/FOC1B bit, an immediate compare match is forced on theWaveform Generation unit. The OC1A/OC1B output is changed according to its COM1x1:0 bits setting. Note thatthe FOC1A/FOC1B bits are implemented as strobes. Therefore it is the value present in the COM1x1:0 bits thatdetermine the effect of the forced compare.

A FOC1A/FOC1B strobe will not generate any interrupt nor will it clear the timer in Clear Timer on Compare match(CTC) mode using OCR1A as TOP.

The FOC1A/FOC1B bits are always read as zero.

• Bit 1:0 – WGM11:0: Waveform Generation ModeCombined with the WGM13:2 bits found in the TCCR1B Register, these bits control the counting sequence of thecounter, the source for maximum (TOP) counter value, and what type of waveform generation to be used, seeTable 16-5. Modes of operation supported by the Timer/Counter unit are: Normal mode (counter), Clear Timer onCompare match (CTC) mode, and three types of Pulse Width Modulation (PWM) modes. (See “Modes of Opera-tion” on page 103.)

Table 17-4. Compare Output Mode, Phase Correct and Phase and Frequency Correct PWM (1)



0 1 WGM13:0 = 9 or 14: Toggle OC1A on Compare Match, OC1B disconnected (normal port operation).For all other WGM13:0 settings, normal port operation, OC1A/OC1B disconnected.

1 0 Clear OC1A/OC1B on compare match when up-counting. Set OC1A/OC1B on compare match when downcounting.

1 1 Set OC1A/OC1B on compare match when up-counting. Clear OC1A/OC1B on compare match when downcounting.


Mode WGM13WGM12(CTC1)

WGM11(PWM11)

WGM10(PWM10)


Update of OCR1x

TOV1 Flag Set on

0 0 0 0 0 Normal 0xFFFF Immediate MAX

1 0 0 0 1 PWM, Phase Correct, 8-bit 0x00FF TOP BOTTOM




Note: 1. The CTC1 and PWM11:0 bit definition names are obsolete. Use the WGM12:0 definitions. However, the functionality and location of these bits are compatible with previous versions of the timer.

17.10.2 TCCR1B – Timer/Counter1 Control Register B

• Bit 7 – ICNC1: Input Capture Noise CancelerSetting this bit (to one) activates the Input Capture Noise Canceler. When the Noise Canceler is activated, theinput from the Input Capture Pin (ICP1) is filtered. The filter function requires four successive equal valued samplesof the ICP1 pin for changing its output. The Input Capture is therefore delayed by four Oscillator cycles when theNoise Canceler is enabled.

• Bit 6 – ICES1: Input Capture Edge SelectThis bit selects which edge on the Input Capture Pin (ICP1) that is used to trigger a capture event. When theICES1 bit is written to zero, a falling (negative) edge is used as trigger, and when the ICES1 bit is written to one, arising (positive) edge will trigger the capture.

When a capture is triggered according to the ICES1 setting, the counter value is copied into the Input Capture Reg-ister (ICR1). The event will also set the Input Capture Flag (ICF1), and this can be used to cause an Input CaptureInterrupt, if this interrupt is enabled.

When the ICR1 is used as TOP value (see description of the WGM13:0 bits located in the TCCR1A and theTCCR1B Register), the ICP1 is disconnected and consequently the Input Capture function is disabled.

• Bit 5 – Reserved BitThis bit is reserved for future use. For ensuring compatibility with future devices, this bit must be written to zerowhen TCCR1B is written.

• Bit 4:3 – WGM13:2: Waveform Generation ModeSee TCCR1A Register description.

4 0 1 0 0 CTC OCR1A Immediate MAX

5 0 1 0 1 Fast PWM, 8-bit 0x00FF BOTTOM TOP



8 1 0 0 0 PWM, Phase and Frequency Correct ICR1 BOTTOM BOTTOM

9 1 0 0 1 PWM, Phase and Frequency Correct OCR1A BOTTOM BOTTOM

10 1 0 1 0 PWM, Phase Correct ICR1 TOP BOTTOM

11 1 0 1 1 PWM, Phase Correct OCR1A TOP BOTTOM

12 1 1 0 0 CTC ICR1 Immediate MAX

13 1 1 0 1 Reserved – – –

14 1 1 1 0 Fast PWM ICR1 BOTTOM TOP

15 1 1 1 1 Fast PWM OCR1A BOTTOM TOP


Mode WGM13WGM12(CTC1)

WGM11(PWM11)

WGM10(PWM10)


Update of OCR1x

TOV1 Flag Set on

Bit 7 6 5 4 3 2 1 0

ICNC1 ICES1 – WGM13 WGM12 CS12 CS11 CS10 TCCR1B




• Bit 2:0 – CS12:0: Clock SelectThe three Clock Select bits select the clock source to be used by the Timer/Counter, see Figure 16-10 and Figure16-11.

If external pin modes are used for the Timer/Counter1, transitions on the T1 pin will clock the counter even if thepin is configured as an output. This feature allows software control of the counting.

17.10.3 TCNT1H and TCNT1L – Timer/Counter1

The two Timer/Counter I/O locations (TCNT1H and TCNT1L, combined TCNT1) give direct access, both for readand for write operations, to the Timer/Counter unit 16-bit counter. To ensure that both the high and low bytes areread and written simultaneously when the CPU accesses these registers, the access is performed using an 8-bittemporary High Byte Register (TEMP). This temporary register is shared by all the other 16-bit registers. See“Accessing 16-bit Registers” on page 94.

Modifying the counter (TCNT1) while the counter is running introduces a risk of missing a compare match betweenTCNT1 and one of the OCR1x Registers.

Writing to the TCNT1 Register blocks (removes) the compare match on the following timer clock for all compareunits.

17.10.4 OCR1AH and OCR1AL – Output Compare Register 1 A

17.10.5 OCR1BH and OCR1BL – Output Compare Register 1 B




0 0 1 clkI/O/1 (No prescaling)





1 1 0 External clock source on T1 pin. Clock on falling edge.

1 1 1 External clock source on T1 pin. Clock on rising edge.

Bit 7 6 5 4 3 2 1 0

TCNT1[15:8] TCNT1H

TCNT1[7:0] TCNT1L



Bit 7 6 5 4 3 2 1 0

OCR1A[15:8] OCR1AH

OCR1A[7:0] OCR1AL



Bit 7 6 5 4 3 2 1 0

OCR1B[15:8] OCR1BH


The Output Compare Registers contain a 16-bit value that is continuously compared with the counter value(TCNT1). A match can be used to generate an output compare interrupt, or to generate a waveform output on theOC1x pin.

The Output Compare Registers are 16-bit in size. To ensure that both the high and low bytes are written simultane-ously when the CPU writes to these registers, the access is performed using an 8-bit temporary High Byte Register(TEMP). This temporary register is shared by all the other 16-bit registers. See “Accessing 16-bit Registers” onpage 94.

17.10.6 ICR1H and ICR1L – Input Capture Register 1

The Input Capture is updated with the counter (TCNT1) value each time an event occurs on the ICP1 pin (oroptionally on the analog comparator output for Timer/Counter1). The Input Capture can be used for defining thecounter TOP value.

The Input Capture Register is 16-bit in size. To ensure that both the high and low bytes are read simultaneouslywhen the CPU accesses these registers, the access is performed using an 8-bit temporary High Byte Register(TEMP). This temporary register is shared by all the other 16-bit registers. See “Accessing 16-bit Registers” onpage 94.

17.10.7 TIMSK – Timer/Counter Interrupt Mask Register(1)

Note: 1. This register contains interrupt control bits for several Timer/Counters, but only Timer1 bits are described in this section. The remaining bits are described in their respective timer sections.

• Bit 5 – TICIE1: Timer/Counter1, Input Capture Interrupt EnableWhen this bit is written to one, and the I-flag in the Status Register is set (interrupts globally enabled), theTimer/Counter1 Input Capture Interrupt is enabled. The corresponding Interrupt Vector (See “Interrupts” on page45.) is executed when the ICF1 Flag, located in TIFR, is set.

• Bit 4 – OCIE1A: Timer/Counter1, Output Compare A Match Interrupt EnableWhen this bit is written to one, and the I-flag in the Status Register is set (interrupts globally enabled), theTimer/Counter1 Output Compare A match interrupt is enabled. The corresponding Interrupt Vector (See “Inter-rupts” on page 45.) is executed when the OCF1A Flag, located in TIFR, is set.

• Bit 3 – OCIE1B: Timer/Counter1, Output Compare B Match Interrupt EnableWhen this bit is written to one, and the I-flag in the Status Register is set (interrupts globally enabled), theTimer/Counter1 Output Compare B match interrupt is enabled. The corresponding Interrupt Vector (See “Inter-rupts” on page 45.) is executed when the OCF1B Flag, located in TIFR, is set.

• Bit 2 – TOIE1: Timer/Counter1, Overflow Interrupt Enable

OCR1B[7:0] OCR1BL



Bit 7 6 5 4 3 2 1 0

ICR1[15:8] ICR1H

ICR1[7:0] ICR1L



Bit 7 6 5 4 3 2 1 0





When this bit is written to one, and the I-flag in the Status Register is set (interrupts globally enabled), theTimer/Counter1 Overflow Interrupt is enabled. The corresponding Interrupt Vector (See “Interrupts” on page 45.) isexecuted when the TOV1 Flag, located in TIFR, is set.


Note: 1. This register contains flag bits for several Timer/Counters, but only Timer1 bits are described in this section. The remaining bits are described in their respective timer sections.

• Bit 5 – ICF1: Timer/Counter1, Input Capture FlagThis flag is set when a capture event occurs on the ICP1 pin. When the Input Capture Register (ICR1) is set by theWGM13:0 to be used as the TOP value, the ICF1 Flag is set when the counter reaches the TOP value.

ICF1 is automatically cleared when the Input Capture Interrupt Vector is executed. Alternatively, ICF1 can becleared by writing a logic one to its bit location.

• Bit 4 – OCF1A: Timer/Counter1, Output Compare A Match FlagThis flag is set in the timer clock cycle after the counter (TCNT1) value matches the Output Compare Register A(OCR1A).

Note that a Forced Output Compare (FOC1A) strobe will not set the OCF1A Flag.

OCF1A is automatically cleared when the Output Compare Match A Interrupt Vector is executed. Alternatively,OCF1A can be cleared by writing a logic one to its bit location.

• Bit 3 – OCF1B: Timer/Counter1, Output Compare B Match FlagThis flag is set in the timer clock cycle after the counter (TCNT1) value matches the Output Compare Register B(OCR1B).

Note that a forced output compare (FOC1B) strobe will not set the OCF1B Flag.

OCF1B is automatically cleared when the Output Compare Match B Interrupt Vector is executed. Alternatively,OCF1B can be cleared by writing a logic one to its bit location.

• Bit 2 – TOV1: Timer/Counter1, Overflow FlagThe setting of this flag is dependent of the WGM13:0 bits setting. In normal and CTC modes, the TOV1 Flag is setwhen the timer overflows. Refer to Table 16-5 on page 114 for the TOV1 Flag behavior when using anotherWGM13:0 bit setting.

TOV1 is automatically cleared when the Timer/Counter1 Overflow interrupt vector is executed. Alternatively, TOV1can be cleared by writing a logic one to its bit location.

Bit 7 6 5 4 3 2 1 0





18. 8-bit Timer/Counter2 with PWM and Asynchronous Operation

18.1 Features• Single Compare unit Counter• Clear Timer on Compare Match (Auto Reload)• Glitch-free, Phase Correct Pulse Width Modulator (PWM)• Frequency Generator• 10-bit Clock Prescaler• Overflow and Compare Match Interrupt Sources (TOV2 and OCF2)• Allows clocking from External 32kHz Watch Crystal Independent of the I/O Clock

18.2 OverviewTimer/Counter2 is a general purpose, single compare unit, 8-bit Timer/Counter module. A simplified block diagramof the 8-bit Timer/Counter is shown in Figure 17-1. For the actual placement of I/O pins, refer to “PinoutATmega32A” on page 2. CPU accessible I/O Registers, including I/O bits and I/O pins, are shown in bold. Thedevice-specific I/O Register and bit locations are listed in the “Register Description” on page 132.

Figure 18-1. 8-bit Timer/Counter Block Diagram

18.2.1 RegistersThe Timer/Counter (TCNT2) and Output Compare Register (OCR2) are 8-bit registers. Interrupt request (shortenas Int.Req.) signals are all visible in the Timer Interrupt Flag Register (TIFR). All interrupts are individually masked

Timer/Counter

DAT

AB

US

=

TCNTn

WaveformGeneration

OCn

= 0

Control Logic

= 0xFF

TOPBOTTOM

count

clear

direction

TOVn(Int.Req.)

OCn(Int.Req.)

Synchronization Unit

OCRn

TCCRn

ASSRnStatus flags

clkI/O

clkASY

Synchronized Status flags

asynchronous modeselect (ASn)

TOSC1

T/COscillator

TOSC2

Prescaler

clkTn

clkI/O


with the Timer Interrupt Mask Register (TIMSK). TIFR and TIMSK are not shown in the figure since these registersare shared by other timer units.

The Timer/Counter can be clocked internally, via the prescaler, or asynchronously clocked from the TOSC1/2 pins,as detailed later in this section. The asynchronous operation is controlled by the Asynchronous Status Register(ASSR). The Clock Select logic block controls which clock source the Timer/Counter uses to increment (or decre-ment) its value. The Timer/Counter is inactive when no clock source is selected. The output from the Clock Selectlogic is referred to as the timer clock (clkT2).

The double buffered Output Compare Register (OCR2) is compared with the Timer/Counter value at all times. Theresult of the compare can be used by the waveform generator to generate a PWM or variable frequency output onthe Output Compare Pin (OC2). See “Output Compare Unit” on page 121. for details. The compare match eventwill also set the Compare Flag (OCF2) which can be used to generate an output compare interrupt request.

18.2.2 DefinitionsMany register and bit references in this document are written in general form. A lower case “n” replaces theTimer/Counter number, in this case 2. However, when using the register or bit defines in a program, the preciseform must be used (that is, TCNT2 for accessing Timer/Counter2 counter value and so on). The definitions in Table17-1 are also used extensively throughout the document.

18.3 Timer/Counter Clock SourcesThe Timer/Counter can be clocked by an internal synchronous or an external asynchronous clock source. Theclock source clkT2 is by default equal to the MCU clock, clkI/O. When the AS2 bit in the ASSR Register is written tologic one, the clock source is taken from the Timer/Counter Oscillator connected to TOSC1 and TOSC2. Fordetails on asynchronous operation, see “‘ASSR – Asynchronous Status Register” on page 135. For details on clocksources and prescaler, see “Timer/Counter Prescaler” on page 132.

18.4 Counter UnitThe main part of the 8-bit Timer/Counter is the programmable bi-directional counter unit. Figure 17-2 shows ablock diagram of the counter and its surrounding environment.




BOTTOM The counter reaches the BOTTOM when it becomes zero (0x00).

MAX The counter reaches its MAXimum when it becomes 0xFF (decimal 255).

TOP The counter reaches the TOP when it becomes equal to the highest value inthe count sequence. The TOP value can be assigned to be the fixed value0xFF (MAX) or the value stored in the OCR2 Register. The assignment isdependent on the mode of operation.

DATA BUS

TCNTn Control Logic

count

TOVn(Int.Req.)

topbottom

direction

clear

TOSC1

T/COscillator

TOSC2

Prescaler

clkI/O

clkTn


count Increment or decrement TCNT2 by 1.

direction Selects between increment and decrement.

clear Clear TCNT2 (set all bits to zero).

clkT2 Timer/Counter clock.

top Signalizes that TCNT2 has reached maximum value.

bottom Signalizes that TCNT2 has reached minimum value (zero).

Depending on the mode of operation used, the counter is cleared, incremented, or decremented at each timerclock (clkT2). clkT2 can be generated from an external or internal clock source, selected by the Clock Select bits(CS22:0). When no clock source is selected (CS22:0 = 0) the timer is stopped. However, the TCNT2 value can beaccessed by the CPU, regardless of whether clkT2 is present or not. A CPU write overrides (has priority over) allcounter clear or count operations.

The counting sequence is determined by the setting of the WGM21 and WGM20 bits located in the Timer/CounterControl Register (TCCR2). There are close connections between how the counter behaves (counts) and howwaveforms are generated on the Output Compare output OC2. For more details about advanced countingsequences and waveform generation, see “Modes of Operation” on page 124.


18.5 Output Compare UnitThe 8-bit comparator continuously compares TCNT2 with the Output Compare Register (OCR2). WheneverTCNT2 equals OCR2, the comparator signals a match. A match will set the Output Compare Flag (OCF2) at thenext timer clock cycle. If enabled (OCIE2 = 1), the Output Compare Flag generates an output compare interrupt.The OCF2 Flag is automatically cleared when the interrupt is executed. Alternatively, the OCF2 Flag can becleared by software by writing a logical one to its I/O bit location. The waveform generator uses the match signal togenerate an output according to operating mode set by the WGM21:0 bits and Compare Output mode (COM21:0)bits. The max and bottom signals are used by the waveform generator for handling the special cases of theextreme values in some modes of operation (“Modes of Operation” on page 124). Figure 17-3 shows a block dia-gram of the output compare unit.



The OCR2 Register is double buffered when using any of the Pulse Width Modulation (PWM) modes. For the nor-mal and Clear Timer on Compare (CTC) modes of operation, the double buffering is disabled. The double bufferingsynchronizes the update of the OCR2 Compare Register to either top or bottom of the counting sequence. Thesynchronization prevents the occurrence of odd-length, non-symmetrical PWM pulses, thereby making the outputglitch-free.

The OCR2 Register access may seem complex, but this is not case. When the double buffering is enabled, theCPU has access to the OCR2 Buffer Register, and if double buffering is disabled the CPU will access the OCR2directly.

18.5.1 Force Output CompareIn non-PWM waveform generation modes, the match output of the comparator can be forced by writing a one tothe Force Output Compare (FOC2) bit. Forcing compare match will not set the OCF2 Flag or reload/clear the timer,but the OC2 pin will be updated as if a real compare match had occurred (the COM21:0 bits settings definewhether the OC2 pin is set, cleared or toggled).

18.5.2 Compare Match Blocking by TCNT2 WriteAll CPU write operations to the TCNT2 Register will block any compare match that occurs in the next timer clockcycle, even when the timer is stopped. This feature allows OCR2 to be initialized to the same value as TCNT2 with-out triggering an interrupt when the Timer/Counter clock is enabled.

18.5.3 Using the Output Compare UnitSince writing TCNT2 in any mode of operation will block all compare matches for one timer clock cycle, there arerisks involved when changing TCNT2 when using the output compare unit, independently of whether theTimer/Counter is running or not. If the value written to TCNT2 equals the OCR2 value, the compare match will bemissed, resulting in incorrect waveform generation. Similarly, do not write the TCNT2 value equal to BOTTOMwhen the counter is downcounting.

The setup of the OC2 should be performed before setting the Data Direction Register for the port pin to output. Theeasiest way of setting the OC2 value is to use the Force Output Compare (FOC2) strobe bit in Normal mode. TheOC2 Register keeps its value even when changing between Waveform Generation modes.

OCFn (Int.Req.)


OCRn

OCxy

DATA BUS

TCNTn

WGMn1:0

Waveform Generator

top

FOCn

COMn1:0

bottom


Be aware that the COM2[1:0] bits are not double buffered together with the compare value. Changing theCOM2[1:0] bits will take effect immediately.

18.6 Compare Match Output UnitThe Compare Output mode (COM21:0) bits have two functions. The Waveform Generator uses the COM2[1:0] bitsfor defining the Output Compare (OC2) state at the next compare match. Also, the COM2[1:0] bits control the OC2pin output source. Figure 17-4 shows a simplified schematic of the logic affected by the COM2[1:0] bit setting. TheI/O Registers, I/O bits, and I/O pins in the figure are shown in bold. Only the parts of the general I/O Port ControlRegisters (DDR and PORT) that are affected by the COM2[1:0] bits are shown. When referring to the OC2 state,the reference is for the internal OC2 Register, not the OC2 pin.


The general I/O port function is overridden by the Output Compare (OC2) from the waveform generator if either ofthe COM2[1:0] bits are set. However, the OC2 pin direction (input or output) is still controlled by the Data DirectionRegister (DDR) for the port pin. The Data Direction Register bit for the OC2 pin (DDR_OC2) must be set as outputbefore the OC2 value is visible on the pin. The port override function is independent of the Waveform Generationmode.

The design of the output compare pin logic allows initialization of the OC2 state before the output is enabled. Notethat some COM2[1:0] bit settings are reserved for certain modes of operation. “Register Description” on page 132

18.6.1 Compare Output Mode and Waveform GenerationThe waveform generator uses the COM2[1:0] bits differently in Normal, CTC, and PWM modes. For all modes, set-ting the COM2[1:0] = 0 tells the Waveform Generator that no action on the OC2 Register is to be performed on thenext compare match. For compare output actions in the non-PWM modes refer to Table 17-3 on page 133. For fastPWM mode, refer to Table 17-4 on page 134, and for phase correct PWM refer to Table 17-5 on page 134.

A change of the COM21:0 bits state will have effect at the first compare match after the bits are written. For non-PWM modes, the action can be forced to have immediate effect by using the FOC2 strobe bits.

18.7 Modes of OperationThe mode of operation, that is, the behavior of the Timer/Counter and the output compare pins, is defined by thecombination of the Waveform Generation mode (WGM2[1:0]) and Compare Output mode (COM2[1:0]) bits. TheCompare Output mode bits do not affect the counting sequence, while the Waveform Generation mode bits do.The COM2[1:0] bits control whether the PWM output generated should be inverted or not (inverted or non-inverted

PORT

DDR

D Q

D Q

OCnPinOCn


COMn1

COMn0

0

1D

ATA

BU

S

FOCn

clkI/O


PWM). For non-PWM modes the COM2[1:0] bits control whether the output should be set, cleared, or toggled at acompare match (See “Compare Match Output Unit” on page 123.).

For detailed timing information refer to “Timer/Counter Timing Diagrams” on page 128.

18.7.1 Normal ModeThe simplest mode of operation is the Normal mode (WGM2[1:0] = 0). In this mode the counting direction is alwaysup (incrementing), and no counter clear is performed. The counter simply overruns when it passes its maximum 8-bit value (TOP = 0xFF) and then restarts from the bottom (0x00). In normal operation the Timer/Counter OverflowFlag (TOV2) will be set in the same timer clock cycle as the TCNT2 becomes zero. The TOV2 Flag in this casebehaves like a ninth bit, except that it is only set, not cleared. However, combined with the timer overflow interruptthat automatically clears the TOV2 Flag, the timer resolution can be increased by software. There are no specialcases to consider in the normal mode, a new counter value can be written anytime.

The Output Compare unit can be used to generate interrupts at some given time. Using the output compare to gen-erate waveforms in normal mode is not recommended, since this will occupy too much of the CPU time.

18.7.2 Clear Timer on Compare Match (CTC) ModeIn Clear Timer on Compare or CTC mode (WGM2[1:0] = 2), the OCR2 Register is used to manipulate the counterresolution. In CTC mode the counter is cleared to zero when the counter value (TCNT2) matches the OCR2. TheOCR2 defines the top value for the counter, hence also its resolution. This mode allows greater control of the com-pare match output frequency. It also simplifies the operation of counting external events.

The timing diagram for the CTC mode is shown in Figure 17-5. The counter value (TCNT2) increases until a com-pare match occurs between TCNT2 and OCR2, and then counter (TCNT2) is cleared.


An interrupt can be generated each time the counter value reaches the TOP value by using the OCF2 Flag. If theinterrupt is enabled, the interrupt handler routine can be used for updating the TOP value. However, changing theTOP to a value close to BOTTOM when the counter is running with none or a low prescaler value must be donewith care since the CTC mode does not have the double buffering feature. If the new value written to OCR2 islower than the current value of TCNT2, the counter will miss the compare match. The counter will then have tocount to its maximum value (0xFF) and wrap around starting at 0x00 before the compare match can occur.

For generating a waveform output in CTC mode, the OC2 output can be set to toggle its logical level on each com-pare match by setting the Compare Output mode bits to toggle mode (COM2[1:0] = 1). The OC2 value will not bevisible on the port pin unless the data direction for the pin is set to output. The waveform generated will have a

TCNTn

OCn(Toggle)


1 4Period 2 3

(COMn1:0 = 1)


maximum frequency of fOC2 = fclk_I/O/2 when OCR2 is set to zero (0x00). The waveform frequency is defined by thefollowing equation:

The N variable represents the prescale factor (1, 8, 32, 64, 128, 256, or 1024).


18.7.3 Fast PWM ModeThe fast Pulse Width Modulation or fast PWM mode (WGM2[1:0] = 3) provides a high frequency PWM waveformgeneration option. The fast PWM differs from the other PWM option by its single-slope operation. The countercounts from BOTTOM to MAX then restarts from BOTTOM. In non-inverting Compare Output mode, the OutputCompare (OC2) is cleared on the compare match between TCNT2 and OCR2, and set at BOTTOM. In invertingCompare Output mode, the output is set on compare match and cleared at BOTTOM. Due to the single-slope oper-ation, the operating frequency of the fast PWM mode can be twice as high as the phase correct PWM mode thatuses dual-slope operation. This high frequency makes the fast PWM mode well suited for power regulation, rectifi-cation, and DAC applications. High frequency allows physically small sized external components (coils,capacitors), and therefore reduces total system cost.

In fast PWM mode, the counter is incremented until the counter value matches the MAX value. The counter is thencleared at the following timer clock cycle. The timing diagram for the fast PWM mode is shown in Figure 17-6. TheTCNT2 value is in the timing diagram shown as a histogram for illustrating the single-slope operation. The diagramincludes non-inverted and inverted PWM outputs. The small horizontal line marks on the TCNT2 slopes representcompare matches between OCR2 and TCNT2.


The Timer/Counter Overflow Flag (TOV2) is set each time the counter reaches MAX. If the interrupt is enabled, theinterrupt handler routine can be used for updating the compare value.

In fast PWM mode, the compare unit allows generation of PWM waveforms on the OC2 pin. Setting the COM2[1:0]bits to 2 will produce a non-inverted PWM and an inverted PWM output can be generated by setting the COM2[1:0]to 3 (see Table 17-4 on page 134). The actual OC2 value will only be visible on the port pin if the data direction forthe port pin is set as output. The PWM waveform is generated by setting (or clearing) the OC2 Register at the com-pare match between OCR2 and TCNT2, and clearing (or setting) the OC2 Register at the timer clock cycle thecounter is cleared (changes from MAX to BOTTOM).

fOCn

fclk_I/O

2 N 1 OCRn+ -----------------------------------------------=

TCNTn

OCRn Update andTOVn Interrupt Flag Set

1Period 2 3

OCn

OCn

(COMn1:0 = 2)

(COMn1:0 = 3)

OCRn Interrupt Flag Set

4 5 6 7




The extreme values for the OCR2 Register represent special cases when generating a PWM waveform output inthe fast PWM mode. If the OCR2 is set equal to BOTTOM, the output will be a narrow spike for each MAX+1 timerclock cycle. Setting the OCR2 equal to MAX will result in a constantly high or low output (depending on the polarityof the output set by the COM21:0 bits.)

A frequency (with 50% duty cycle) waveform output in fast PWM mode can be achieved by setting OC2 to toggleits logical level on each compare match (COM2[1:0] = 1). The waveform generated will have a maximum frequencyof foc2 = fclk_I/O/2 when OCR2 is set to zero. This feature is similar to the OC2 toggle in CTC mode, except the dou-ble buffer feature of the output compare unit is enabled in the fast PWM mode.

18.7.4 Phase Correct PWM ModeThe phase correct PWM mode (WGM2[1:0] = 1) provides a high resolution phase correct PWM waveform genera-tion option. The phase correct PWM mode is based on a dual-slope operation. The counter counts repeatedly fromBOTTOM to MAX and then from MAX to BOTTOM. In non-inverting Compare Output mode, the Output Compare(OC2) is cleared on the compare match between TCNT2 and OCR2 while upcounting, and set on the comparematch while downcounting. In inverting Output Compare mode, the operation is inverted. The dual-slope operationhas lower maximum operation frequency than single slope operation. However, due to the symmetric feature of thedual-slope PWM modes, these modes are preferred for motor control applications.

The PWM resolution for the phase correct PWM mode is fixed to 8 bits. In phase correct PWM mode the counter isincremented until the counter value matches MAX. When the counter reaches MAX, it changes the count direction.The TCNT2 value will be equal to MAX for one timer clock cycle. The timing diagram for the phase correct PWMmode is shown on Figure 17-7. The TCNT2 value is in the timing diagram shown as a histogram for illustrating thedual-slope operation. The diagram includes non-inverted and inverted PWM outputs. The small horizontal linemarks on the TCNT2 slopes represent compare matches between OCR2 and TCNT2.


The Timer/Counter Overflow Flag (TOV2) is set each time the counter reaches BOTTOM. The Interrupt Flag can beused to generate an interrupt each time the counter reaches the BOTTOM value.

fOCnPWM

fclk_I/O

N 256------------------=

TOVn Interrupt Flag Set


1 2 3

TCNTn

Period

OCn

OCn

(COMn1:0 = 2)

(COMn1:0 = 3)

OCRn Update


In phase correct PWM mode, the compare unit allows generation of PWM waveforms on the OC2 pin. Setting theCOM21:0 bits to 2 will produce a non-inverted PWM. An inverted PWM output can be generated by setting theCOM21:0 to 3 (see Table 17-5 on page 134). The actual OC2 value will only be visible on the port pin if the datadirection for the port pin is set as output. The PWM waveform is generated by clearing (or setting) the OC2 Regis-ter at the compare match between OCR2 and TCNT2 when the counter increments, and setting (or clearing) theOC2 Register at compare match between OCR2 and TCNT2 when the counter decrements. The PWM frequencyfor the output when using phase correct PWM can be calculated by the following equation:


The extreme values for the OCR2 Register represent special cases when generating a PWM waveform output inthe phase correct PWM mode. If the OCR2 is set equal to BOTTOM, the output will be continuously low and if setequal to MAX the output will be continuously high for non-inverted PWM mode. For inverted PWM the output willhave the opposite logic values.

At the very start of period 2 in Figure 17-7 OCn has a transition from high to low even though there is no CompareMatch. The point of this transition is to guarantee symmetry around BOTTOM. THere are two cases that give atransition without Compare Match.

• OCR2A chages its value from MAX, like in Figure 17-7. When the OCR2A value is MAX the OCn pin value is the same as the result of a down-counting Compare Match. To ensure symmetry around BOTTOM the OCn value at MAX must correspond to the result of an up-counting Compare Match.

• The timer starts counting from a value higher than the one in OCR2A, and for that reason misses the Compare Match and hence the OCn change that would have happened on the way up.

18.8 Timer/Counter Timing DiagramsThe following figures show the Timer/Counter in Synchronous mode, and the timer clock (clkT2) is therefore shownas a clock enable signal. In Asynchronous mode, clkI/O should be replaced by the Timer/Counter Oscillator clock.The figures include information on when Interrupt Flags are set. Figure 17-8 contains timing data for basicTimer/Counter operation. The figure shows the count sequence close to the MAX value in all modes other thanphase correct PWM mode.



fOCnPCPWM

fclk_I/O

N 510------------------=

clkTn(clkI/O/1)

TOVn

clkI/O




Figure 17-10 shows the setting of OCF2 in all modes except CTC mode.

Figure 18-10. Timer/Counter Timing Diagram, Setting of OCF2, with Prescaler (fclk_I/O/8)

Figure 17-11 shows the setting of OCF2 and the clearing of TCNT2 in CTC mode.

TOVn


clkI/O

clkTn(clkI/O/8)

OCFn

OCRn

TCNTn

OCRn Value

OCRn - 1 OCRn OCRn + 1 OCRn + 2

clkI/O

clkTn(clkI/O/8)


Figure 18-11. Timer/Counter Timing Diagram, Clear Timer on Compare Match Mode, with Prescaler (fclk_I/O/8)

18.9 Asynchronous Operation of the Timer/CounterWhen Timer/Counter2 operates asynchronously, some considerations must be taken.

• Warning: When switching between asynchronous and synchronous clocking of Timer/Counter2, the Timer Registers TCNT2, OCR2, and TCCR2 might be corrupted. A safe procedure for switching clock source is:

1. Disable the Timer/Counter2 interrupts by clearing OCIE2 and TOIE2.

2. Select clock source by setting AS2 as appropriate.

3. Write new values to TCNT2, OCR2, and TCCR2.

4. To switch to asynchronous operation: Wait for TCN2UB, OCR2UB, and TCR2UB.

5. Clear the Timer/Counter2 Interrupt Flags.

6. Enable interrupts, if needed.

• The Oscillator is optimized for use with a 32.768 kHz watch crystal. Applying an external clock to the TOSC1 pin may result in incorrect Timer/Counter2 operation. The CPU main clock frequency must be more than four times the Oscillator frequency.

• When writing to one of the registers TCNT2, OCR2, or TCCR2, the value is transferred to a temporary register, and latched after two positive edges on TOSC1. The user should not write a new value before the contents of the temporary register have been transferred to its destination. Each of the three mentioned registers have their individual temporary register, which means for example that writing to TCNT2 does not disturb an OCR2 write in progress. To detect that a transfer to the destination register has taken place, the Asynchronous Status Register – ASSR has been implemented.

• When entering Power-save or Extended Standby mode after having written to TCNT2, OCR2, or TCCR2, the user must wait until the written register has been updated if Timer/Counter2 is used to wake up the device. Otherwise, the MCU will enter sleep mode before the changes are effective. This is particularly important if the Output Compare2 interrupt is used to wake up the device, since the output compare function is disabled during writing to OCR2 or TCNT2. If the write cycle is not finished, and the MCU enters sleep mode before the OCR2UB bit returns to zero, the device will never receive a compare match interrupt, and the MCU will not wake up.

• If Timer/Counter2 is used to wake the device up from Power-save or Extended Standby mode, precautions must be taken if the user wants to re-enter one of these modes: The interrupt logic needs one TOSC1 cycle to be reset. If the time between wake-up and re-entering sleep mode is less than one TOSC1 cycle, the interrupt will not occur, and the device will fail to wake up. If the user is in doubt whether the time before re-entering Power-

OCFn

OCRn

TCNTn(CTC)

TOP


clkI/O

clkTn(clkI/O/8)


save or Extended Standby mode is sufficient, the following algorithm can be used to ensure that one TOSC1 cycle has elapsed:

1. Write a value to TCCR2, TCNT2, or OCR2.

2. Wait until the corresponding Update Busy Flag in ASSR returns to zero.

3. Enter Power-save or Extended Standby mode.

• When the asynchronous operation is selected, the 32.768kHz Oscillator for Timer/Counter2 is always running, except in Power-down and Standby modes. After a Power-up Reset or wake-up from Power-down or Standby mode, the user should be aware of the fact that this Oscillator might take as long as one second to stabilize. The user is advised to wait for at least one second before using Timer/Counter2 after power-up or wake-up from Power-down or Standby mode. The contents of all Timer/Counter2 Registers must be considered lost after a wake-up from Power-down or Standby mode due to unstable clock signal upon start-up, no matter whether the Oscillator is in use or a clock signal is applied to the TOSC1 pin.

• Description of wake up from Power-save or Extended Standby mode when the timer is clocked asynchronously: When the interrupt condition is met, the wake up process is started on the following cycle of the timer clock, that is, the timer is always advanced by at least one before the processor can read the counter value. After wake-up, the MCU is halted for four cycles, it executes the interrupt routine, and resumes execution from the instruction following SLEEP.

• Reading of the TCNT2 Register shortly after wake-up from Power-save may give an incorrect result. Since TCNT2 is clocked on the asynchronous TOSC clock, reading TCNT2 must be done through a register synchronized to the internal I/O clock domain. Synchronization takes place for every rising TOSC1 edge. When waking up from Power-save mode, and the I/O clock (clkI/O) again becomes active, TCNT2 will read as the previous value (before entering sleep) until the next rising TOSC1 edge. The phase of the TOSC clock after waking up from Power-save mode is essentially unpredictable, as it depends on the wake-up time. The recommended procedure for reading TCNT2 is thus as follows:

1. Write any value to either of the registers OCR2 or TCCR2.

2. Wait for the corresponding Update Busy Flag to be cleared.

3. Read TCNT2.

• During asynchronous operation, the synchronization of the Interrupt Flags for the asynchronous timer takes three processor cycles plus one timer cycle. The timer is therefore advanced by at least one before the processor can read the timer value causing the setting of the Interrupt Flag. The output compare pin is changed on the timer clock and is not synchronized to the processor clock.


18.10 Timer/Counter Prescaler

Figure 18-12. Prescaler for Timer/Counter2

The clock source for Timer/Counter2 is named clkT2S. clkT2S is by default connected to the main system I/O clockclkIO. By setting the AS2 bit in ASSR, Timer/Counter2 is asynchronously clocked from the TOSC1 pin. Thisenables use of Timer/Counter2 as a Real Time Counter (RTC). When AS2 is set, pins TOSC1 and TOSC2 are dis-connected from Port C. A crystal can then be connected between the TOSC1 and TOSC2 pins to serve as anindependent clock source for Timer/Counter2. The Oscillator is optimized for use with a 32.768kHz crystal. Apply-ing an external clock source to TOSC1 is not recommended.

For Timer/Counter2, the possible prescaled selections are: clkT2S/8, clkT2S/32, clkT2S/64, clkT2S/128, clkT2S/256, andclkT2S/1024. Additionally, clkT2S as well as 0 (stop) may be selected. Setting the PSR2 bit in SFIOR resets the pres-caler. This allows the user to operate with a predictable prescaler.


18.11.1 TCCR2 – Timer/Counter Control Register

• Bit 7 – FOC2: Force Output CompareThe FOC2 bit is only active when the WGM bits specify a non-PWM mode. However, for ensuring compatibility withfuture devices, this bit must be set to zero when TCCR2 is written when operating in PWM mode. When writing alogical one to the FOC2 bit, an immediate compare match is forced on the waveform generation unit. The OC2 out-put is changed according to its COM21:0 bits setting. Note that the FOC2 bit is implemented as a strobe. Thereforeit is the value present in the COM21:0 bits that determines the effect of the forced compare.

A FOC2 strobe will not generate any interrupt, nor will it clear the timer in CTC mode using OCR2 as TOP.

The FOC2 bit is always read as zero.

10-BIT T/C PRESCALER

TIMER/COUNTER2 CLOCK SOURCE

clkI/O clkT2S

TOSC1

AS2

CS20CS21CS22

clk T

2S/8

clk T

2S/6

4

clk T

2S/1

28

clk T

2S/1

024

clk T

2S/2

56

clk T

2S/3

2

0PSR2

Clear

clkT2

Bit 7 6 5 4 3 2 1 0

FOC2 WGM20 COM21 COM20 WGM21 CS22 CS21 CS20 TCCR2

Read/Write W R/W R/W R/W R/W R/W R/W R/W



• Bit 6, 3 – WGM2[1:0]: Waveform Generation ModeThese bits control the counting sequence of the counter, the source for the maximum (TOP) counter value, andwhat type of waveform generation to be used. Modes of operation supported by the Timer/Counter unit are: Normalmode, Clear Timer on Compare match (CTC) mode, and two types of Pulse Width Modulation (PWM) modes. SeeTable 17-2 and “Modes of Operation” on page 124.

Note: 1. The CTC2 and PWM2 bit definition names are now obsolete. Use the WGM21:0 definitions. However, the function-ality and location of these bits are compatible with previous versions of the timer.

• Bit 5:4 – COM2[1:0]: Compare Match Output ModeThese bits control the Output Compare pin (OC2) behavior. If one or both of the COM21:0 bits are set, the OC2output overrides the normal port functionality of the I/O pin it is connected to. However, note that the Data DirectionRegister (DDR) bit corresponding to OC2 pin must be set in order to enable the output driver.

When OC2 is connected to the pin, the function of the COM21:0 bits depends on the WGM21:0 bit setting. Table17-3 shows the COM21:0 bit functionality when the WGM21:0 bits are set to a normal or CTC mode (non-PWM).

Table 17-4 shows the COM2[1:0] bit functionality when the WGM21:0 bits are set to fast PWM mode.

Note: 1. A special case occurs when OCR2 equals TOP and COM21 is set. In this case, the compare match is ignored, but the set or clear is done at TOP. See “Fast PWM Mode” on page 125 for more details.

Table 17-5 shows the COM2[1:0] bit functionality when the WGM21:0 bits are set to phase correct PWM mode


ModeWGM21(CTC2)

WGM20(PWM2)


Update ofOCR2

TOV2 FlagSet on

0 0 0 Normal 0xFF Immediate MAX

1 0 1 PWM, Phase Correct 0xFF TOP BOTTOM

2 1 0 CTC OCR2 Immediate MAX

3 1 1 Fast PWM 0xFF BOTTOM MAX

Table 18-3. Compare Output Mode, non-PWM Mode



0 1 Toggle OC2 on compare match

1 0 Clear OC2 on compare match

1 1 Set OC2 on compare match

Table 18-4. Compare Output Mode, Fast PWM Mode(1)



0 1 Reserved

1 0 Clear OC2 on compare match, set OC2 at BOTTOM,

(non-inverting mode)

1 1 Set OC2 on compare match, clear OC2 at BOTTOM,(inverting mode)


.

Note: 1. A special case occurs when OCR2 equals TOP and COM21 is set. In this case, the compare match is ignored, but the set or clear is done at TOP. See “Phase Correct PWM Mode” on page 127 for more details.

• Bit 2:0 – CS2[2:0]: Clock SelectThe three Clock Select bits select the clock source to be used by the Timer/Counter, see Table 17-6.

18.11.2 TCNT2 – Timer/Counter Register

The Timer/Counter Register gives direct access, both for read and write operations, to the Timer/Counter unit 8-bitcounter. Writing to the TCNT2 Register blocks (removes) the compare match on the following timer clock. Modify-ing the counter (TCNT2) while the counter is running, introduces a risk of missing a compare match betweenTCNT2 and the OCR2 Register.

18.11.3 OCR2 – Output Compare Register

The Output Compare Register contains an 8-bit value that is continuously compared with the counter value(TCNT2). A match can be used to generate an output compare interrupt, or to generate a waveform output on theOC2 pin.

Table 18-5. Compare Output Mode, Phase Correct PWM Mode(1)



0 1 Reserved

1 0 Clear OC2 on compare match when up-counting. Set OC2 on compare match when downcounting.

1 1 Set OC2 on compare match when up-counting. Clear OC2 on compare match when downcounting.




0 0 1 clkT2S/(No prescaling)

0 1 0 clkT2S/8 (From prescaler)






Bit 7 6 5 4 3 2 1 0

TCNT2[7:0] TCNT2



Bit 7 6 5 4 3 2 1 0

OCR2[7:0] OCR2




18.11.4 ‘ASSR – Asynchronous Status Register

• Bit 3 – AS2: Asynchronous Timer/Counter2When AS2 is written to zero, Timer/Counter 2 is clocked from the I/O clock, clkI/O. When AS2 is written to one,Timer/Counter2 is clocked from a Crystal Oscillator connected to the Timer Oscillator 1 (TOSC1) pin. When thevalue of AS2 is changed, the contents of TCNT2, OCR2, and TCCR2 might be corrupted.

• Bit 2 – TCN2UB: Timer/Counter2 Update BusyWhen Timer/Counter2 operates asynchronously and TCNT2 is written, this bit becomes set. When TCNT2 hasbeen updated from the temporary storage register, this bit is cleared by hardware. A logical zero in this bit indicatesthat TCNT2 is ready to be updated with a new value.

• Bit 1 – OCR2UB: Output Compare Register2 Update BusyWhen Timer/Counter2 operates asynchronously and OCR2 is written, this bit becomes set. When OCR2 has beenupdated from the temporary storage register, this bit is cleared by hardware. A logical zero in this bit indicates thatOCR2 is ready to be updated with a new value.

• Bit 0 – TCR2UB: Timer/Counter Control Register2 Update BusyWhen Timer/Counter2 operates asynchronously and TCCR2 is written, this bit becomes set. When TCCR2 hasbeen updated from the temporary storage register, this bit is cleared by hardware. A logical zero in this bit indicatesthat TCCR2 is ready to be updated with a new value.

If a write is performed to any of the three Timer/Counter2 Registers while its update busy flag is set, the updatedvalue might get corrupted and cause an unintentional interrupt to occur.

The mechanisms for reading TCNT2, OCR2, and TCCR2 are different. When reading TCNT2, the actual timervalue is read. When reading OCR2 or TCCR2, the value in the temporary storage register is read.

18.11.5 TIMSK – Timer/Counter Interrupt Mask Register

• Bit 7 – OCIE2: Timer/Counter2 Output Compare Match Interrupt EnableWhen the OCIE2 bit is written to one and the I-bit in the Status Register is set (one), the Timer/Counter2 CompareMatch interrupt is enabled. The corresponding interrupt is executed if a compare match in Timer/Counter2 occurs,that is, when the OCF2 bit is set in the Timer/Counter Interrupt Flag Register – TIFR.

• Bit 6 – TOIE2: Timer/Counter2 Overflow Interrupt EnableWhen the TOIE2 bit is written to one and the I-bit in the Status Register is set (one), the Timer/Counter2 Overflowinterrupt is enabled. The corresponding interrupt is executed if an overflow in Timer/Counter2 occurs, that is, whenthe TOV2 bit is set in the Timer/Counter Interrupt Flag Register – TIFR.

Bit 7 6 5 4 3 2 1 0

– – – – AS2 TCN2UB OCR2UB TCR2UB ASSR

Read/Write R R R R R/W R R R


Bit 7 6 5 4 3 2 1 0






• Bit 7 – OCF2: Output Compare Flag 2The OCF2 bit is set (one) when a compare match occurs between the Timer/Counter2 and the data in OCR2 –Output Compare Register2. OCF2 is cleared by hardware when executing the corresponding interrupt handlingvector. Alternatively, OCF2 is cleared by writing a logic one to the flag. When the I-bit in SREG, OCIE2(Timer/Counter2 Compare match Interrupt Enable), and OCF2 are set (one), the Timer/Counter2 Compare matchInterrupt is executed.

• Bit 6 – TOV2: Timer/Counter2 Overflow FlagThe TOV2 bit is set (one) when an overflow occurs in Timer/Counter2. TOV2 is cleared by hardware when execut-ing the corresponding interrupt handling vector. Alternatively, TOV2 is cleared by writing a logic one to the flag.When the SREG I-bit, TOIE2 (Timer/Counter2 Overflow Interrupt Enable), and TOV2 are set (one), theTimer/Counter2 Overflow interrupt is executed. In PWM mode, this bit is set when Timer/Counter2 changes count-ing direction at $00.


• Bit 1 – PSR2: Prescaler Reset Timer/Counter2When this bit is written to one, the Timer/Counter2 prescaler will be reset. The bit will be cleared by hardware afterthe operation is performed. Writing a zero to this bit will have no effect. This bit will always be read as zero ifTimer/Counter2 is clocked by the internal CPU clock. If this bit is written when Timer/Counter2 is operating in asyn-chronous mode, the bit will remain one until the prescaler has been reset.

Bit 7 6 5 4 3 2 1 0




Bit 7 6 5 4 3 2 1 0





19. SPI – Serial Peripheral Interface

19.1 Features• Full-duplex, Three-wire Synchronous Data Transfer• Master or Slave Operation• LSB First or MSB First Data Transfer• Seven Programmable Bit Rates• End of Transmission Interrupt Flag• Write Collision Flag Protection• Wake-up from Idle Mode• Double Speed (CK/2) Master SPI Mode

19.2 OverviewThe Serial Peripheral Interface (SPI) allows high-speed synchronous data transfer between the ATmega32A andperipheral devices or between several AVR devices.

Figure 19-1. SPI Block Diagram(1)

Note: 1. Refer to Figure 1-1 on page 2, and Table 12-6 on page 59 for SPI pin placement.

The interconnection between Master and Slave CPUs with SPI is shown in Figure 18-2. The system consists of twoShift Registers, and a Master clock generator. The SPI Master initiates the communication cycle when pulling lowthe Slave Select SS pin of the desired Slave. Master and Slave prepare the data to be sent in their respective Shift

SP

I2X

SP

I2X

DIVIDER/2/4/8/16/32/64/128


Registers, and the Master generates the required clock pulses on the SCK line to interchange data. Data is alwaysshifted from Master to Slave on the Master Out – Slave In, MOSI, line, and from Slave to Master on the Master In –Slave Out, MISO, line. After each data packet, the Master will synchronize the Slave by pulling high the SlaveSelect, SS, line.

When configured as a Master, the SPI interface has no automatic control of the SS line. This must be handled byuser software before communication can start. When this is done, writing a byte to the SPI Data Register starts theSPI clock generator, and the hardware shifts the eight bits into the Slave. After shifting one byte, the SPI clock gen-erator stops, setting the end of Transmission Flag (SPIF). If the SPI Interrupt Enable bit (SPIE) in the SPCRRegister is set, an interrupt is requested. The Master may continue to shift the next byte by writing it into SPDR, orsignal the end of packet by pulling high the Slave Select, SS line. The last incoming byte will be kept in the BufferRegister for later use.

When configured as a Slave, the SPI interface will remain sleeping with MISO tri-stated as long as the SS pin isdriven high. In this state, software may update the contents of the SPI Data Register, SPDR, but the data will notbe shifted out by incoming clock pulses on the SCK pin until the SS pin is driven low. As one byte has been com-pletely shifted, the end of Transmission Flag, SPIF is set. If the SPI Interrupt Enable bit, SPIE, in the SPCRRegister is set, an interrupt is requested. The Slave may continue to place new data to be sent into SPDR beforereading the incoming data. The last incoming byte will be kept in the Buffer Register for later use.

Figure 19-2. SPI Master-slave Interconnection

The system is single buffered in the transmit direction and double buffered in the receive direction. This means thatbytes to be transmitted cannot be written to the SPI Data Register before the entire shift cycle is completed. Whenreceiving data, however, a received character must be read from the SPI Data Register before the next characterhas been completely shifted in. Otherwise, the first byte is lost.

In SPI Slave mode, the control logic will sample the incoming signal of the SCK pin. To ensure correct sampling ofthe clock signal, the minimum low and high periods should be:

Low periods: longer than 2 CPU clock cycles.

High periods: longer than 2 CPU clock cycles.

When the SPI is enabled, the data direction of the MOSI, MISO, SCK, and SS pins is overridden according toTable 18-1. For more details on automatic port overrides, refer to “Alternate Port Functions” on page 55.

Table 19-1. SPI Pin Overrides

Pin Direction, Master SPI Direction, Slave SPI

MOSI User Defined Input

MISO Input User Defined

SCK User Defined Input

SS User Defined Input

MSB MASTER LSB

8 BIT SHIFT REGISTER

MSB SLAVE LSB

8 BIT SHIFT REGISTERMISO

MOSI

SPICLOCK GENERATOR

SCK

SS

MISO

MOSI

SCK

SS

SHIFTENABLE


Note: See “Alternate Functions of Port B” on page 59 for a detailed description of how to define the direction of the userdefined SPI pins.

The following code examples show how to initialize the SPI as a master and how to perform a simple transmission.DDR_SPI in the examples must be replaced by the actual Data Direction Register controlling the SPI pins.DD_MOSI, DD_MISO and DD_SCK must be replaced by the actual data direction bits for these pins. For exampleif MOSI is placed on pin PB5, replace DD_MOSI with DDB5 and DDR_SPI with DDRB.



SPI_MasterInit:

; Set MOSI and SCK output, all others input

ldi r17,(1<<DD_MOSI)|(1<<DD_SCK)

out DDR_SPI,r17

; Enable SPI, Master, set clock rate fck/16

ldi r17,(1<<SPE)|(1<<MSTR)|(1<<SPR0)

out SPCR,r17

ret

SPI_MasterTransmit:

; Start transmission of data (r16)

out SPDR,r16

Wait_Transmit:

; Wait for transmission complete

sbis SPSR,SPIF

rjmp Wait_Transmit

ret

C Code Example(1)

void SPI_MasterInit(void)

{

/* Set MOSI and SCK output, all others input */

DDR_SPI = (1<<DD_MOSI)|(1<<DD_SCK);

/* Enable SPI, Master, set clock rate fck/16 */

SPCR = (1<<SPE)|(1<<MSTR)|(1<<SPR0);

}

void SPI_MasterTransmit(char cData)

{

/* Start transmission */

SPDR = cData;

/* Wait for transmission complete */

while(!(SPSR & (1<<SPIF)))

;

}


The following code examples show how to initialize the SPI as a Slave and how to perform a simple reception.



SPI_SlaveInit:

; Set MISO output, all others input

ldi r17,(1<<DD_MISO)

out DDR_SPI,r17

; Enable SPI

ldi r17,(1<<SPE)

out SPCR,r17

ret

SPI_SlaveReceive:

; Wait for reception complete

sbis SPSR,SPIF

rjmp SPI_SlaveReceive

; Read received data and return

in r16,SPDR

ret

C Code Example(1)

void SPI_SlaveInit(void)

{

/* Set MISO output, all others input */

DDR_SPI = (1<<DD_MISO);

/* Enable SPI */

SPCR = (1<<SPE);

}

char SPI_SlaveReceive(void)

{

/* Wait for reception complete */

while(!(SPSR & (1<<SPIF)))

;

/* Return data register */

return SPDR;

}


19.3 SS Pin Functionality

19.3.1 Slave ModeWhen the SPI is configured as a Slave, the Slave Select (SS) pin is always input. When SS is held low, the SPI isactivated, and MISO becomes an output if configured so by the user. All other pins are inputs. When SS is drivenhigh, all pins are inputs except MISO which can be user configured as an output, and the SPI is passive, whichmeans that it will not receive incoming data. Note that the SPI logic will be reset once the SS pin is driven high.

The SS pin is useful for packet/byte synchronization to keep the slave bit counter synchronous with the masterclock generator. When the SS pin is driven high, the SPI Slave will immediately reset the send and receive logic,and drop any partially received data in the Shift Register.

19.3.2 Master ModeWhen the SPI is configured as a Master (MSTR in SPCR is set), the user can determine the direction of the SS pin.

If SS is configured as an output, the pin is a general output pin which does not affect the SPI system. Typically, thepin will be driving the SS pin of the SPI Slave.

If SS is configured as an input, it must be held high to ensure Master SPI operation. If the SS pin is driven low byperipheral circuitry when the SPI is configured as a Master with the SS pin defined as an input, the SPI systeminterprets this as another master selecting the SPI as a slave and starting to send data to it. To avoid bus conten-tion, the SPI system takes the following actions:

1. The MSTR bit in SPCR is cleared and the SPI system becomes a slave. As a result of the SPI becoming a slave, the MOSI and SCK pins become inputs.

2. The SPIF Flag in SPSR is set, and if the SPI interrupt is enabled, and the I-bit in SREG is set, the interrupt routine will be executed.

Thus, when interrupt-driven SPI transmission is used in master mode, and there exists a possibility that SS isdriven low, the interrupt should always check that the MSTR bit is still set. If the MSTR bit has been cleared by aslave select, it must be set by the user to re-enable SPI master mode.

19.3.3 SPCR – SPI Control Register

• Bit 7 – SPIE: SPI Interrupt EnableThis bit causes the SPI interrupt to be executed if SPIF bit in the SPSR Register is set and the if the global interruptenable bit in SREG is set.

• Bit 6 – SPE: SPI EnableWhen the SPE bit is written to one, the SPI is enabled. This bit must be set to enable any SPI operations.

• Bit 5 – DORD: Data OrderWhen the DORD bit is written to one, the LSB of the data word is transmitted first.

When the DORD bit is written to zero, the MSB of the data word is transmitted first.

Bit 7 6 5 4 3 2 1 0

SPIE SPE DORD MSTR CPOL CPHA SPR1 SPR0 SPCR




• Bit 4 – MSTR: Master/Slave SelectThis bit selects Master SPI mode when written to one, and Slave SPI mode when written logic zero. If SS is config-ured as an input and is driven low while MSTR is set, MSTR will be cleared, and SPIF in SPSR will become set.The user will then have to set MSTR to re-enable SPI Master mode.

• Bit 3 – CPOL: Clock PolarityWhen this bit is written to one, SCK is high when idle. When CPOL is written to zero, SCK is low when idle. Referto Figure 18-3 and Figure 18-4 for an example. The CPOL functionality is summarized below:

• Bit 2 – CPHA: Clock PhaseThe settings of the Clock Phase bit (CPHA) determine if data is sampled on the leading (first) or trailing (last) edgeof SCK. Refer to Figure 18-3 and Figure 18-4 for an example. The CPHA functionality is summarized below:

• Bits 1, 0 – SPR1, SPR0: SPI Clock Rate Select 1 and 0These two bits control the SCK rate of the device configured as a Master. SPR1 and SPR0 have no effect on theSlave. The relationship between SCK and the Oscillator Clock frequency fosc is shown in the following table:

19.3.4 SPSR – SPI Status Register

• Bit 7 – SPIF: SPI Interrupt FlagWhen a serial transfer is complete, the SPIF Flag is set. An interrupt is generated if SPIE in SPCR is set and globalinterrupts are enabled. If SS is an input and is driven low when the SPI is in Master mode, this will also set the SPIF

Table 19-2. CPOL Functionality

CPOL Leading Edge Trailing Edge

0 Rising Falling

1 Falling Rising

Table 19-3. CPHA Functionality

CPHA Leading Edge Trailing Edge

0 Sample Setup

1 Setup Sample

Table 19-4. Relationship Between SCK and the Oscillator Frequency

SPI2X SPR1 SPR0 SCK Frequency

0 0 0 fosc/4

0 0 1 fosc/16

0 1 0 fosc/64

0 1 1 fosc/128

1 0 0 fosc/2

1 0 1 fosc/8

1 1 0 fosc/32

1 1 1 fosc/64

Bit 7 6 5 4 3 2 1 0

SPIF WCOL – – – – – SPI2X SPSR

Read/Write R R R R R R R R/W



Flag. SPIF is cleared by hardware when executing the corresponding interrupt handling vector. Alternatively, theSPIF bit is cleared by first reading the SPI Status Register with SPIF set, then accessing the SPI Data Register(SPDR).

• Bit 6 – WCOL: Write COLlision FlagThe WCOL bit is set if the SPI Data Register (SPDR) is written during a data transfer. The WCOL bit (and the SPIFbit) are cleared by first reading the SPI Status Register with WCOL set, and then accessing the SPI Data Register.

• Bit 5:1 – Reserved BitsThese bits are reserved bits in the ATmega32A and will always read as zero.

• Bit 0 – SPI2X: Double SPI Speed BitWhen this bit is written logic one the SPI speed (SCK Frequency) will be doubled when the SPI is in Master mode(see Table 18-4). This means that the minimum SCK period will be two CPU clock periods. When the SPI is config-ured as Slave, the SPI is only guaranteed to work at fosc/4 or lower.

The SPI interface on the ATmega32A is also used for program memory and EEPROM downloading or uploading.See page 279 for SPI Serial Programming and Verification.

19.3.5 SPDR – SPI Data Register

The SPI Data Register is a read/write register used for data transfer between the Register File and the SPI ShiftRegister. Writing to the register initiates data transmission. Reading the register causes the Shift Register Receivebuffer to be read.

19.4 Data ModesThere are four combinations of SCK phase and polarity with respect to serial data, which are determined by controlbits CPHA and CPOL. The SPI data transfer formats are shown in Figure 18-3 and Figure 18-4. Data bits areshifted out and latched in on opposite edges of the SCK signal, ensuring sufficient time for data signals to stabilize.This is clearly seen by summarizing Table 18-2 and Table 18-3, as done below:

Bit 7 6 5 4 3 2 1 0

MSB LSB SPDR


Initial Value X X X X X X X X Undefined

Table 19-5. CPOL and CPHA Functionality

Leading Edge Trailing Edge SPI Mode

CPOL = 0, CPHA = 0 Sample (Rising) Setup (Falling) 0

CPOL = 0, CPHA = 1 Setup (Rising) Sample (Falling) 1

CPOL = 1, CPHA = 0 Sample (Falling) Setup (Rising) 2

CPOL = 1, CPHA = 1 Setup (Falling) Sample (Rising) 3


Figure 19-3. SPI Transfer Format with CPHA = 0

Figure 19-4. SPI Transfer Format with CPHA = 1

Bit 1Bit 6

LSBMSB

SCK (CPOL = 0)mode 0

SAMPLE IMOSI/MISO

CHANGE 0MOSI PIN

CHANGE 0MISO PIN


SS

MSBLSB

Bit 6Bit 1

Bit 5Bit 2

Bit 4Bit 3

Bit 3Bit 4

Bit 2Bit 5

MSB first (DORD = 0)LSB first (DORD = 1)


SAMPLE IMOSI/MISO

CHANGE 0MOSI PIN

CHANGE 0MISO PIN


SS

MSBLSB

Bit 6Bit 1

Bit 5Bit 2

Bit 4Bit 3

Bit 3Bit 4

Bit 2Bit 5

Bit 1Bit 6

LSBMSB

MSB first (DORD = 0)LSB first (DORD = 1)


20. USART

20.1 Features• Full Duplex Operation (Independent Serial Receive and Transmit Registers)• Asynchronous or Synchronous Operation• Master or Slave Clocked Synchronous Operation• High Resolution Baud Rate Generator• Supports Serial Frames with 5, 6, 7, 8, or 9 Data Bits and 1 or 2 Stop Bits• Odd or Even Parity Generation and Parity Check Supported by Hardware• Data OverRun Detection• Framing Error Detection• Noise Filtering Includes False Start Bit Detection and Digital Low Pass Filter• Three Separate Interrupts on TX Complete, TX Data Register Empty, and RX Complete• Multi-processor Communication Mode• Double Speed Asynchronous Communication Mode

20.2 OverviewThe Universal Synchronous and Asynchronous serial Receiver and Transmitter (USART) is a highly flexible serialcommunication device. A simplified block diagram of the USART transmitter is shown in Figure 19-1. CPU accessi-ble I/O Registers and I/O pins are shown in bold.


Figure 20-1. USART Block Diagram(1)

Note: 1. Refer to Figure 1-1 on page 2, Table 12-14 on page 66, and Table 12-8 on page 61 for USART pin placement.

The dashed boxes in the block diagram separate the three main parts of the USART (listed from the top): ClockGenerator, Transmitter and Receiver. Control Registers are shared by all units. The clock generation logic consistsof synchronization logic for external clock input used by synchronous slave operation, and the baud rate generator.The XCK (Transfer Clock) pin is only used by Synchronous Transfer mode. The Transmitter consists of a singlewrite buffer, a serial Shift Register, parity generator and control logic for handling different serial frame formats.The write buffer allows a continuous transfer of data without any delay between frames. The Receiver is the mostcomplex part of the USART module due to its clock and data recovery units. The recovery units are used for asyn-chronous data reception. In addition to the recovery units, the receiver includes a parity checker, control logic, aShift Register and a two level receive buffer (UDR). The receiver supports the same frame formats as the transmit-ter, and can detect frame error, data overrun and parity errors.

20.2.1 AVR USART vs. AVR UART – CompatibilityThe USART is fully compatible with the AVR UART regarding:

• Bit locations inside all USART Registers

• Baud Rate Generation

• Transmitter Operation

• Transmit Buffer Functionality

PARITYGENERATOR

UBRR[H:L]

UDR (Transmit)

UCSRA UCSRB UCSRC

BAUD RATE GENERATOR

TRANSMIT SHIFT REGISTER

RECEIVE SHIFT REGISTER RxD

TxDPIN

CONTROL

UDR (Receive)

PINCONTROL

XCK

DATARECOVERY

CLOCKRECOVERY

PINCONTROL

TXCONTROL

RXCONTROL

PARITYCHECKER

DA

TA

BU

S

OSC

SYNC LOGIC

Clock Generator

Transmitter

Receiver


• Receiver Operation

However, the receive buffering has two improvements that will affect the compatibility in some special cases:

• A second Buffer Register has been added. The two Buffer Registers operate as a circular FIFO buffer. Therefore the UDR must only be read once for each incoming data! More important is the fact that the Error Flags (FE and DOR) and the 9th data bit (RXB8) are buffered with the data in the receive buffer. Therefore the status bits must always be read before the UDR Register is read. Otherwise the error status will be lost since the buffer state is lost.

• The receiver Shift Register can now act as a third buffer level. This is done by allowing the received data to remain in the serial Shift Register (see Figure 19-1) if the Buffer Registers are full, until a new start bit is detected. The USART is therefore more resistant to Data OverRun (DOR) error conditions.

The following control bits have changed name, but have same functionality and register location:

• CHR9 is changed to UCSZ2

• OR is changed to DOR

20.3 Clock GenerationThe clock generation logic generates the base clock for the Transmitter and Receiver. The USART supports fourmodes of clock operation: Normal Asynchronous, Double Speed Asynchronous, Master Synchronous and SlaveSynchronous mode. The UMSEL bit in USART Control and Status Register C (UCSRC) selects between asyn-chronous and synchronous operation. Double Speed (Asynchronous mode only) is controlled by the U2X found inthe UCSRA Register. When using Synchronous mode (UMSEL = 1), the Data Direction Register for the XCK pin(DDR_XCK) controls whether the clock source is internal (Master mode) or external (Slave mode). The XCK pin isonly active when using Synchronous mode.

Figure 19-2 shows a block diagram of the clock generation logic.

Figure 20-2. Clock Generation Logic, Block Diagram

Signal description:

txclk Transmitter clock (Internal Signal).

rxclk Receiver base clock (Internal Signal).

xcki Input from XCK pin (Internal Signal). Used for synchronous slave operation.

xcko Clock output to XCK pin (Internal Signal). Used for synchronous masteroperation.

PrescalingDown-Counter

/ 2

UBRR

/ 4 / 2

fosc

UBRR+1

SyncRegister

OSC

XCKPin

txclk

U2X

UMSEL

DDR_XCK

0

1

0

1

xcki

xcko

DDR_XCKrxclk

0

1

1

0

EdgeDetector

UCPOL


fosc XTAL pin frequency (System Clock).

20.3.1 Internal Clock Generation – The Baud Rate GeneratorInternal clock generation is used for the asynchronous and the synchronous master modes of operation. Thedescription in this section refers to Figure 19-2.

The USART Baud Rate Register (UBRR) and the down-counter connected to it function as a programmable pres-caler or baud rate generator. The down-counter, running at system clock (fosc), is loaded with the UBRR valueeach time the counter has counted down to zero or when the UBRRL Register is written. A clock is generated eachtime the counter reaches zero. This clock is the baud rate generator clock output (= fosc/(UBRR+1)). The Trans-mitter divides the baud rate generator clock output by 2, 8 or 16 depending on mode. The baud rate generatoroutput is used directly by the receiver’s clock and data recovery units. However, the recovery units use a statemachine that uses 2, 8 or 16 states depending on mode set by the state of the UMSEL, U2X and DDR_XCK bits.

Table 19-1 contains equations for calculating the baud rate (in bits per second) and for calculating the UBRR valuefor each mode of operation using an internally generated clock source.

Note: 1. The baud rate is defined to be the transfer rate in bit per second (bps).

BAUD Baud rate (in bits per second, bps)

fOSC System Oscillator clock frequency

UBRR Contents of the UBRRH and UBRRL Registers, (0 - 4095)

Some examples of UBRR values for some system clock frequencies are found in Table 19-9 (see page 172).

20.3.2 Double Speed Operation (U2X)The transfer rate can be doubled by setting the U2X bit in UCSRA. Setting this bit only has effect for the asynchro-nous operation. Set this bit to zero when using synchronous operation.

Setting this bit will reduce the divisor of the baud rate divider from 16 to 8, effectively doubling the transfer rate forasynchronous communication. Note however that the receiver will in this case only use half the number of samples(reduced from 16 to 8) for data sampling and clock recovery, and therefore a more accurate baud rate setting andsystem clock are required when this mode is used. For the Transmitter, there are no downsides.

20.3.3 External ClockExternal clocking is used by the synchronous slave modes of operation. The description in this section refers toFigure 19-2 for details.

External clock input from the XCK pin is sampled by a synchronization register to minimize the chance of meta-sta-bility. The output from the synchronization register must then pass through an edge detector before it can be used

Table 20-1. Equations for Calculating Baud Rate Register Setting

Operating ModeEquation for Calculating

Baud Rate(1)Equation for Calculating

UBRR Value

Asynchronous Normal Mode (U2X = 0)

Asynchronous Double Speed Mode (U2X = 1)

Synchronous Master Mode

BAUDfOSC

16 UBRR 1+ ---------------------------------------= UBRR

fOSC

16BAUD------------------------ 1–=

BAUDfOSC

8 UBRR 1+ -----------------------------------= UBRR

fOSC

8BAUD-------------------- 1–=

BAUDfOSC

2 UBRR 1+ -----------------------------------= UBRR

fOSC

2BAUD-------------------- 1–=


by the Transmitter and receiver. This process introduces a two CPU clock period delay and therefore the maximumexternal XCK clock frequency is limited by the following equation:

Note that fosc depends on the stability of the system clock source. It is therefore recommended to add some marginto avoid possible loss of data due to frequency variations.

20.3.4 Synchronous Clock OperationWhen Synchronous mode is used (UMSEL = 1), the XCK pin will be used as either clock input (Slave) or clock out-put (Master). The dependency between the clock edges and data sampling or data change is the same. The basicprinciple is that data input (on RxD) is sampled at the opposite XCK clock edge of the edge the data output (TxD) ischanged.

Figure 20-3. Synchronous Mode XCK Timing.

The UCPOL bit UCRSC selects which XCK clock edge is used for data sampling and which is used for datachange. As Figure 19-3 shows, when UCPOL is zero the data will be changed at rising XCK edge and sampled atfalling XCK edge. If UCPOL is set, the data will be changed at falling XCK edge and sampled at rising XCK edge.

20.4 Frame FormatsA serial frame is defined to be one character of data bits with synchronization bits (start and stop bits), and option-ally a parity bit for error checking. The USART accepts all 30 combinations of the following as valid frame formats:

• 1 start bit

• 5, 6, 7, 8, or 9 data bits

• no, even or odd parity bit

• 1 or 2 stop bits

A frame starts with the start bit followed by the least significant data bit. Then the next data bits, up to a total ofnine, are succeeding, ending with the most significant bit. If enabled, the parity bit is inserted after the data bits,before the stop bits. When a complete frame is transmitted, it can be directly followed by a new frame, or the com-munication line can be set to an idle (high) state. Figure 19-4 illustrates the possible combinations of the frameformats. Bits inside brackets are optional.

Figure 20-4. Frame Formats

fXCK

fOSC

4-----------

RxD / TxD

XCK

RxD / TxD

XCKUCPOL = 0

UCPOL = 1

Sample

Sample

10 2 3 4 [5] [6] [7] [8] [P]St Sp1 [Sp2] (St / IDLE)(IDLE)

FRAME


St Start bit, always low.

(n) Data bits (0 to 8).

P Parity bit. Can be odd or even.

Sp Stop bit, always high.

IDLE No transfers on the communication line (RxD or TxD). An IDLE line must behigh.

The frame format used by the USART is set by the UCSZ2:0, UPM1:0, and USBS bits in UCSRB and UCSRC. TheReceiver and Transmitter use the same setting. Note that changing the setting of any of these bits will corrupt allongoing communication for both the Receiver and Transmitter.

The USART Character SiZe (UCSZ2:0) bits select the number of data bits in the frame. The USART Parity mode(UPM1:0) bits enable and set the type of parity bit. The selection between one or two stop bits is done by theUSART Stop Bit Select (USBS) bit. The receiver ignores the second stop bit. An FE (Frame Error) will thereforeonly be detected in the cases where the first stop bit is zero.

20.4.1 Parity Bit CalculationThe parity bit is calculated by doing an exclusive-or of all the data bits. If odd parity is used, the result of the exclu-sive or is inverted. The relation between the parity bit and data bits is as follows::

Peven Parity bit using even parity

Podd Parity bit using odd parity

dn Data bit n of the character

If used, the parity bit is located between the last data bit and first stop bit of a serial frame.

20.5 USART InitializationThe USART has to be initialized before any communication can take place. The initialization process normally con-sists of setting the baud rate, setting frame format and enabling the Transmitter or the Receiver depending on theusage. For interrupt driven USART operation, the Global Interrupt Flag should be cleared (and interrupts globallydisabled) when doing the initialization.

Before doing a re-initialization with changed baud rate or frame format, be sure that there are no ongoing transmis-sions during the period the registers are changed. The TXC Flag can be used to check that the Transmitter hascompleted all transfers, and the RXC Flag can be used to check that there are no unread data in the receive buffer.Note that the TXC Flag must be cleared before each transmission (before UDR is written) if it is used for thispurpose.

The following simple USART initialization code examples show one assembly and one C function that are equal infunctionality. The examples assume asynchronous operation using polling (no interrupts enabled) and a fixedframe format. The baud rate is given as a function parameter. For the assembly code, the baud rate parameter isassumed to be stored in the r17:r16 registers. When the function writes to the UCSRC Register, the URSEL bit(MSB) must be set due to the sharing of I/O location by UBRRH and UCSRC.

Peven dn 1– d3 d2 d1 d0 0Podd

dn 1– d3 d2 d1 d0 1

==



More advanced initialization routines can be made that include frame format as parameters, disable interrupts andso on. However, many applications use a fixed setting of the Baud and Control Registers, and for these types ofapplications the initialization code can be placed directly in the main routine, or be combined with initialization codefor other I/O modules.

20.6 Data Transmission – The USART TransmitterThe USART Transmitter is enabled by setting the Transmit Enable (TXEN) bit in the UCSRB Register. When theTransmitter is enabled, the normal port operation of the TxD pin is overridden by the USART and given the functionas the transmitter’s serial output. The baud rate, mode of operation and frame format must be set up once beforedoing any transmissions. If synchronous operation is used, the clock on the XCK pin will be overridden and usedas transmission clock.

20.6.1 Sending Frames with 5 to 8 Data BitA data transmission is initiated by loading the transmit buffer with the data to be transmitted. The CPU can load thetransmit buffer by writing to the UDR I/O location. The buffered data in the transmit buffer will be moved to the ShiftRegister when the Shift Register is ready to send a new frame. The Shift Register is loaded with new data if it is inidle state (no ongoing transmission) or immediately after the last stop bit of the previous frame is transmitted. Whenthe Shift Register is loaded with new data, it will transfer one complete frame at the rate given by the Baud Regis-ter, U2X bit or by XCK depending on mode of operation.

The following code examples show a simple USART transmit function based on polling of the Data Register Empty(UDRE) Flag. When using frames with less than eight bits, the most significant bits written to the UDR are ignored.


USART_Init:

; Set baud rate

out UBRRH, r17

out UBRRL, r16

; Enable receiver and transmitter

ldi r16, (1<<RXEN)|(1<<TXEN)

out UCSRB,r16

; Set frame format: 8data, 2stop bit

ldi r16, (1<<URSEL)|(1<<USBS)|(3<<UCSZ0)

out UCSRC,r16

ret

C Code Example(1)

void USART_Init( unsigned int baud )

{

/* Set baud rate */

UBRRH = (unsigned char)(baud>>8);

UBRRL = (unsigned char)baud;

/* Enable receiver and transmitter */

UCSRB = (1<<RXEN)|(1<<TXEN);

/* Set frame format: 8data, 2stop bit */

UCSRC = (1<<URSEL)|(1<<USBS)|(3<<UCSZ0);

}


The USART has to be initialized before the function can be used. For the assembly code, the data to be sent isassumed to be stored in Register R16.


The function simply waits for the transmit buffer to be empty by checking the UDRE Flag, before loading it with newdata to be transmitted. If the Data Register Empty Interrupt is utilized, the interrupt routine writes the data into thebuffer.

20.6.2 Sending Frames with 9 Data BitIf 9-bit characters are used (UCSZ = 7), the ninth bit must be written to the TXB8 bit in UCSRB before the low byteof the character is written to UDR. The following code examples show a transmit function that handles 9-bit charac-ters. For the assembly code, the data to be sent is assumed to be stored in Registers R17:R16.


USART_Transmit:

; Wait for empty transmit buffer

sbis UCSRA,UDRE

rjmp USART_Transmit

; Put data (r16) into buffer, sends the data

out UDR,r16

ret

C Code Example(1)

void USART_Transmit( unsigned char data )

{

/* Wait for empty transmit buffer */

while ( !( UCSRA & (1<<UDRE)) )

;

/* Put data into buffer, sends the data */

UDR = data;

}


Note: 1. These transmit functions are written to be general functions. They can be optimized if the contents of the UCSRB is static (that is, only the TXB8 bit of the UCSRB Register is used after initialization).

The ninth bit can be used for indicating an address frame when using multi processor communication mode or forother protocol handling as for example synchronization.

20.6.3 Transmitter Flags and InterruptsThe USART transmitter has two flags that indicate its state: USART Data Register Empty (UDRE) and TransmitComplete (TXC). Both flags can be used for generating interrupts.

The Data Register Empty (UDRE) Flag indicates whether the transmit buffer is ready to receive new data. This bitis set when the transmit buffer is empty, and cleared when the transmit buffer contains data to be transmitted thathas not yet been moved into the Shift Register. For compatibility with future devices, always write this bit to zerowhen writing the UCSRA Register.

When the Data Register empty Interrupt Enable (UDRIE) bit in UCSRB is written to one, the USART Data RegisterEmpty Interrupt will be executed as long as UDRE is set (provided that global interrupts are enabled). UDRE iscleared by writing UDR. When interrupt-driven data transmission is used, the Data Register Empty Interrupt routinemust either write new data to UDR in order to clear UDRE or disable the Data Register empty Interrupt, otherwisea new interrupt will occur once the interrupt routine terminates.

The Transmit Complete (TXC) Flag bit is set one when the entire frame in the transmit Shift Register has beenshifted out and there are no new data currently present in the transmit buffer. The TXC Flag bit is automaticallycleared when a transmit complete interrupt is executed, or it can be cleared by writing a one to its bit location. The


USART_Transmit:

; Wait for empty transmit buffer

sbis UCSRA,UDRE

rjmp USART_Transmit

; Copy 9th bit from r17 to TXB8

cbi UCSRB,TXB8

sbrc r17,0

sbi UCSRB,TXB8

; Put LSB data (r16) into buffer, sends the data

out UDR,r16

ret

C Code Example(1)

void USART_Transmit( unsigned int data )

{

/* Wait for empty transmit buffer */

while ( !( UCSRA & (1<<UDRE))) )

;

/* Copy 9th bit to TXB8 */

UCSRB &= ~(1<<TXB8);

if ( data & 0x0100 )

UCSRB |= (1<<TXB8);

/* Put data into buffer, sends the data */

UDR = data;

}


TXC Flag is useful in half-duplex communication interfaces (like the RS485 standard), where a transmittingapplication must enter receive mode and free the communication bus immediately after completing thetransmission.

When the Transmit Compete Interrupt Enable (TXCIE) bit in UCSRB is set, the USART Transmit Complete Inter-rupt will be executed when the TXC Flag becomes set (provided that global interrupts are enabled). When thetransmit complete interrupt is used, the interrupt handling routine does not have to clear the TXC Flag, this is doneautomatically when the interrupt is executed.

20.6.4 Parity GeneratorThe parity generator calculates the parity bit for the serial frame data. When parity bit is enabled (UPM1 = 1), thetransmitter control logic inserts the parity bit between the last data bit and the first stop bit of the frame that is sent.

20.6.5 Disabling the TransmitterThe disabling of the transmitter (setting the TXEN to zero) will not become effective until ongoing and pendingtransmissions are completed, that is, when the transmit Shift Register and transmit Buffer Register do not containdata to be transmitted. When disabled, the transmitter will no longer override the TxD pin.

20.7 Data Reception – The USART ReceiverThe USART Receiver is enabled by writing the Receive Enable (RXEN) bit in the UCSRB Register to one. Whenthe receiver is enabled, the normal pin operation of the RxD pin is overridden by the USART and given the functionas the receiver’s serial input. The baud rate, mode of operation and frame format must be set up once before anyserial reception can be done. If synchronous operation is used, the clock on the XCK pin will be used as transferclock.

20.7.1 Receiving Frames with 5 to 8 Data BitsThe receiver starts data reception when it detects a valid start bit. Each bit that follows the start bit will be sampledat the baud rate or XCK clock, and shifted into the receive Shift Register until the first stop bit of a frame isreceived. A second stop bit will be ignored by the receiver. When the first stop bit is received, that is, a completeserial frame is present in the receive Shift Register, the contents of the Shift Register will be moved into the receivebuffer. The receive buffer can then be read by reading the UDR I/O location.

The following code example shows a simple USART receive function based on polling of the Receive Complete(RXC) Flag. When using frames with less than eight bits the most significant bits of the data read from the UDR willbe masked to zero. The USART has to be initialized before the function can be used.



The function simply waits for data to be present in the receive buffer by checking the RXC Flag, before reading thebuffer and returning the value.

20.7.2 Receiving Frames with 9 DatabitsIf 9 bit characters are used (UCSZ=7) the ninth bit must be read from the RXB8 bit in UCSRB before reading thelow bits from the UDR. This rule applies to the FE, DOR and PE Status Flags as well. Read status from UCSRA,then data from UDR. Reading the UDR I/O location will change the state of the receive buffer FIFO and conse-quently the TXB8, FE, DOR and PE bits, which all are stored in the FIFO, will change.

The following code example shows a simple USART receive function that handles both 9-bit characters and thestatus bits.


USART_Receive:

; Wait for data to be received

sbis UCSRA, RXC

rjmp USART_Receive

; Get and return received data from buffer

in r16, UDR

ret

C Code Example(1)

unsigned char USART_Receive( void )

{

/* Wait for data to be received */

while ( !(UCSRA & (1<<RXC)) )

;

/* Get and return received data from buffer */

return UDR;

}



The receive function example reads all the I/O Registers into the Register File before any computation is done.This gives an optimal receive buffer utilization since the buffer location read will be free to accept new data as earlyas possible.


USART_Receive:

; Wait for data to be received

sbis UCSRA, RXC

rjmp USART_Receive

; Get status and 9th bit, then data from buffer

in r18, UCSRA

in r17, UCSRB

in r16, UDR

; If error, return -1

andi r18,(1<<FE)|(1<<DOR)|(1<<PE)

breq USART_ReceiveNoError

ldi r17, HIGH(-1)

ldi r16, LOW(-1)

USART_ReceiveNoError:

; Filter the 9th bit, then return

lsr r17

andi r17, 0x01

ret

C Code Example(1)

unsigned int USART_Receive( void )

{

unsigned char status, resh, resl;

/* Wait for data to be received */

while ( !(UCSRA & (1<<RXC)) )

;

/* Get status and 9th bit, then data */

/* from buffer */

status = UCSRA;

resh = UCSRB;

resl = UDR;

/* If error, return -1 */

if ( status & (1<<FE)|(1<<DOR)|(1<<PE) )

return -1;

/* Filter the 9th bit, then return */

resh = (resh >> 1) & 0x01;

return ((resh << 8) | resl);

}


20.7.3 Receive Compete Flag and InterruptThe USART Receiver has one flag that indicates the receiver state.

The Receive Complete (RXC) Flag indicates if there are unread data present in the receive buffer. This flag is onewhen unread data exist in the receive buffer, and zero when the receive buffer is empty (that is, does not containany unread data). If the receiver is disabled (RXEN = 0), the receive buffer will be flushed and consequently theRXC bit will become zero.

When the Receive Complete Interrupt Enable (RXCIE) in UCSRB is set, the USART Receive Complete Interruptwill be executed as long as the RXC Flag is set (provided that global interrupts are enabled). When interrupt-drivendata reception is used, the receive complete routine must read the received data from UDR in order to clear theRXC Flag, otherwise a new interrupt will occur once the interrupt routine terminates.

20.7.4 Receiver Error FlagsThe USART Receiver has three Error Flags: Frame Error (FE), Data OverRun (DOR) and Parity Error (PE). All canbe accessed by reading UCSRA. Common for the Error Flags is that they are located in the receive buffer togetherwith the frame for which they indicate the error status. Due to the buffering of the Error Flags, the UCSRA must beread before the receive buffer (UDR), since reading the UDR I/O location changes the buffer read location. Anotherequality for the Error Flags is that they can not be altered by software doing a write to the flag location. However, allflags must be set to zero when the UCSRA is written for upward compatibility of future USART implementations.None of the Error Flags can generate interrupts.

The Frame Error (FE) Flag indicates the state of the first stop bit of the next readable frame stored in the receivebuffer. The FE Flag is zero when the stop bit was correctly read (as one), and the FE Flag will be one when thestop bit was incorrect (zero). This flag can be used for detecting out-of-sync conditions, detecting break conditionsand protocol handling. The FE Flag is not affected by the setting of the USBS bit in UCSRC since the receiverignores all, except for the first, stop bits. For compatibility with future devices, always set this bit to zero when writ-ing to UCSRA.

The Data OverRun (DOR) Flag indicates data loss due to a receiver buffer full condition. A Data OverRun occurswhen the receive buffer is full (two characters), it is a new character waiting in the receive Shift Register, and a newstart bit is detected. If the DOR Flag is set there was one or more serial frame lost between the frame last readfrom UDR, and the next frame read from UDR. For compatibility with future devices, always write this bit to zerowhen writing to UCSRA. The DOR Flag is cleared when the frame received was successfully moved from the ShiftRegister to the receive buffer.

The Parity Error (PE) Flag indicates that the next frame in the receive buffer had a parity error when received. Ifparity check is not enabled the PE bit will always be read zero. For compatibility with future devices, always set thisbit to zero when writing to UCSRA. For more details see “Parity Bit Calculation” on page 153 and “Parity Checker”on page 160.

20.7.5 Parity CheckerThe Parity Checker is active when the high USART Parity mode (UPM1) bit is set. Type of parity check to be per-formed (odd or even) is selected by the UPM0 bit. When enabled, the parity checker calculates the parity of thedata bits in incoming frames and compares the result with the parity bit from the serial frame. The result of thecheck is stored in the receive buffer together with the received data and stop bits. The Parity Error (PE) Flag canthen be read by software to check if the frame had a parity error.

The PE bit is set if the next character that can be read from the receive buffer had a parity error when received andthe parity checking was enabled at that point (UPM1 = 1). This bit is valid until the receive buffer (UDR) is read.

20.7.6 Disabling the ReceiverIn contrast to the Transmitter, disabling of the Receiver will be immediate. Data from ongoing receptions will there-fore be lost. When disabled (that is, the RXEN is set to zero) the Receiver will no longer override the normal


function of the RxD port pin. The receiver buffer FIFO will be flushed when the receiver is disabled. Remaining datain the buffer will be lost.

20.7.7 Flushing the Receive BufferThe receiver buffer FIFO will be flushed when the Receiver is disabled, that is, the buffer will be emptied of its con-tents. Unread data will be lost. If the buffer has to be flushed during normal operation, due to for instance an errorcondition, read the UDR I/O location until the RXC Flag is cleared. The following code example shows how to flushthe receive buffer.


20.8 Asynchronous Data ReceptionThe USART includes a clock recovery and a data recovery unit for handling asynchronous data reception. Theclock recovery logic is used for synchronizing the internally generated baud rate clock to the incoming asynchro-nous serial frames at the RxD pin. The data recovery logic samples and low pass filters each incoming bit, therebyimproving the noise immunity of the receiver. The asynchronous reception operational range depends on the accu-racy of the internal baud rate clock, the rate of the incoming frames, and the frame size in number of bits.

20.8.1 Asynchronous Clock RecoveryThe clock recovery logic synchronizes internal clock to the incoming serial frames. Figure 19-5 illustrates the sam-pling process of the start bit of an incoming frame. The sample rate is 16 times the baud rate for Normal mode, and8 times the baud rate for Double Speed mode. The horizontal arrows illustrate the synchronization variation due tothe sampling process. Note the larger time variation when using the double speed mode (U2X = 1) of operation.Samples denoted zero are samples done when the RxD line is idle (that is, no communication activity).

Figure 20-5. Start Bit Sampling

When the clock recovery logic detects a high (idle) to low (start) transition on the RxD line, the start bit detectionsequence is initiated. Let sample 1 denote the first zero-sample as shown in the figure. The clock recovery logicthen uses samples 8, 9, and 10 for Normal mode, and samples 4, 5, and 6 for Double Speed mode (indicated withsample numbers inside boxes on the figure), to decide if a valid start bit is received. If two or more of these three


USART_Flush:

sbis UCSRA, RXC

ret

in r16, UDR

rjmp USART_Flush

C Code Example(1)

void USART_Flush( void )

{

unsigned char dummy;

while ( UCSRA & (1<<RXC) ) dummy = UDR;

}

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 1 2

STARTIDLE

00

BIT 0

3

1 2 3 4 5 6 7 8 1 20

RxD

Sample(U2X = 0)

Sample(U2X = 1)


samples have logical high levels (the majority wins), the start bit is rejected as a noise spike and the receiver startslooking for the next high to low-transition. If however, a valid start bit is detected, the clock recovery logic is syn-chronized and the data recovery can begin. The synchronization process is repeated for each start bit.

20.8.2 Asynchronous Data RecoveryWhen the receiver clock is synchronized to the start bit, the data recovery can begin. The data recovery unit uses astate machine that has 16 states for each bit in normal mode and 8 states for each bit in Double Speed mode. Fig-ure 19-6 shows the sampling of the data bits and the parity bit. Each of the samples is given a number that is equalto the state of the recovery unit.

Figure 20-6. Sampling of Data and Parity Bit

The decision of the logic level of the received bit is taken by doing a majority voting of the logic value to the threesamples in the center of the received bit. The center samples are emphasized on the figure by having the samplenumber inside boxes. The majority voting process is done as follows: If two or all three samples have high levels,the received bit is registered to be a logic 1. If two or all three samples have low levels, the received bit is regis-tered to be a logic 0. This majority voting process acts as a low pass filter for the incoming signal on the RxD pin.The recovery process is then repeated until a complete frame is received. Including the first stop bit. Note that thereceiver only uses the first stop bit of a frame.

Figure 19-7 shows the sampling of the stop bit and the earliest possible beginning of the start bit of the next frame.

Figure 20-7. Stop Bit Sampling and Next Start Bit Sampling

The same majority voting is done to the stop bit as done for the other bits in the frame. If the stop bit is registered tohave a logic 0 value, the Frame Error (FE) Flag will be set.

A new high to low transition indicating the start bit of a new frame can come right after the last of the bits used formajority voting. For Normal Speed mode, the first low level sample can be at point marked (A) in Figure 19-7. ForDouble Speed mode the first low level must be delayed to (B). (C) marks a stop bit of full length. The early start bitdetection influences the operational range of the receiver.

20.8.3 Asynchronous Operational RangeThe operational range of the receiver is dependent on the mismatch between the received bit rate and the inter-nally generated baud rate. If the Transmitter is sending frames at too fast or too slow bit rates, or the internally

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 1

BIT n

1 2 3 4 5 6 7 8 1

RxD

Sample(U2X = 0)

Sample(U2X = 1)

1 2 3 4 5 6 7 8 9 10 0/1 0/1 0/1

STOP 1

1 2 3 4 5 6 0/1

RxD

Sample(U2X = 0)

Sample(U2X = 1)

(A) (B) (C)


generated baud rate of the receiver does not have a similar (see Table 19-2) base frequency, the receiver will notbe able to synchronize the frames to the start bit.

The following equations can be used to calculate the ratio of the incoming data rate and internal receiver baud rate.

D Sum of character size and parity size (D = 5 to 10 bit)

S Samples per bit. S = 16 for Normal Speed mode and S = 8 for Double Speed mode.

SF First sample number used for majority voting. SF = 8 for Normal Speed and SF = 4 for Double Speed mode.

SM Middle sample number used for majority voting. SM = 9 for Normal Speed and SM = 5 for Double Speed mode.

Rslow is the ratio of the slowest incoming data rate that can be accepted in relation to thereceiver baud rate. Rfast is the ratio of the fastest incoming data rate that can beaccepted in relation to the receiver baud rate.

Table 19-2 and Table 19-3 list the maximum receiver baud rate error that can be tolerated. Note that NormalSpeed mode has higher toleration of baud rate variations.

Table 20-2. Recommended Maximum Receiver Baud Rate Error for Normal Speed Mode (U2X = 0)

D# (Data+Parity Bit) Rslow (%) Rfast(%)

Max Total Error (%)

Recommended Max Receiver Error (%)

5 93.20 106.67 +6.67/-6.8 ±3.0

6 94.12 105.79 +5.79/-5.88 ±2.5

7 94.81 105.11 +5.11/-5.19 ±2.0

8 95.36 104.58 +4.58/-4.54 ±2.0

9 95.81 104.14 +4.14/-4.19 ±1.5

10 96.17 103.78 +3.78/-3.83 ±1.5

Table 20-3. Recommended Maximum Receiver Baud Rate Error for Double Speed Mode (U2X = 1)

D# (Data+Parity Bit) Rslow (%) Rfast (%)

Max Total Error (%)

Recommended Max Receiver Error (%)

5 94.12 105.66 +5.66/-5.88 ±2.5

6 94.92 104.92 +4.92/-5.08 ±2.0

7 95.52 104.35 +4.35/-4.48 ±1.5

8 96.00 103.90 +3.90/-4.00 ±1.5

9 96.39 103.53 +3.53/-3.61 ±1.5

10 96.70 103.23 +3.23/-3.30 ±1.0

RslowD 1+ S

S 1– D S SF+ +-------------------------------------------=

RfastD 2+ S

D 1+ S SM+-----------------------------------=


The recommendations of the maximum receiver baud rate error was made under the assumption that the receiverand transmitter equally divides the maximum total error.

There are two possible sources for the receivers baud rate error. The receiver’s system clock (XTAL) will alwayshave some minor instability over the supply voltage range and the temperature range. When using a crystal to gen-erate the system clock, this is rarely a problem, but for a resonator the system clock may differ more than 2%depending of the resonators tolerance. The second source for the error is more controllable. The baud rate gener-ator can not always do an exact division of the system frequency to get the baud rate wanted. In this case anUBRR value that gives an acceptable low error can be used if possible.

20.9 Multi-processor Communication ModeSetting the Multi-processor Communication mode (MPCM) bit in UCSRA enables a filtering function of incomingframes received by the USART Receiver. Frames that do not contain address information will be ignored and notput into the receive buffer. This effectively reduces the number of incoming frames that has to be handled by theCPU, in a system with multiple MCUs that communicate via the same serial bus. The Transmitter is unaffected bythe MPCM setting, but has to be used differently when it is a part of a system utilizing the Multi-processor Commu-nication mode.

If the receiver is set up to receive frames that contain 5 to 8 data bits, then the first stop bit indicates if the framecontains data or address information. If the receiver is set up for frames with nine data bits, then the ninth bit(RXB8) is used for identifying address and data frames. When the frame type bit (the first stop or the ninth bit) isone, the frame contains an address. When the frame type bit is zero the frame is a data frame.

The Multi-processor Communication mode enables several slave MCUs to receive data from a master MCU. Thisis done by first decoding an address frame to find out which MCU has been addressed. If a particular Slave MCUhas been addressed, it will receive the following data frames as normal, while the other slave MCUs will ignore thereceived frames until another address frame is received.

20.9.1 Using MPCMFor an MCU to act as a master MCU, it can use a 9-bit character frame format (UCSZ = 7). The ninth bit (TXB8)must be set when an address frame (TXB8 = 1) or cleared when a data frame (TXB = 0) is being transmitted. Theslave MCUs must in this case be set to use a 9-bit character frame format.

The following procedure should be used to exchange data in Multi-processor Communication mode:

1. All slave MCUs are in Multi-processor Communication mode (MPCM in UCSRA is set).

2. The Master MCU sends an address frame, and all slaves receive and read this frame. In the Slave MCUs, the RXC Flag in UCSRA will be set as normal.

3. Each Slave MCU reads the UDR Register and determines if it has been selected. If so, it clears the MPCM bit in UCSRA, otherwise it waits for the next address byte and keeps the MPCM setting.

4. The addressed MCU will receive all data frames until a new address frame is received. The other slave MCUs, which still have the MPCM bit set, will ignore the data frames.

5. When the last data frame is received by the addressed MCU, the addressed MCU sets the MPCM bit and waits for a new address frame from Master. The process then repeats from 2.

Using any of the 5-bit to 8-bit character frame formats is possible, but impractical since the receiver must changebetween using n and n+1 character frame formats. This makes full-duplex operation difficult since the transmitterand receiver uses the same character size setting. If 5-bit to 8-bit character frames are used, the transmitter mustbe set to use two stop bit (USBS = 1) since the first stop bit is used for indicating the frame type.

Do not use Read-Modify-Write instructions (SBI and CBI) to set or clear the MPCM bit. The MPCM bit shares thesame I/O location as the TXC Flag and this might accidentally be cleared when using SBI or CBI instructions.


20.10 Accessing UBRRH/ UCSRC RegistersThe UBRRH Register shares the same I/O location as the UCSRC Register. Therefore some special considerationmust be taken when accessing this I/O location.

20.10.1 Write AccessWhen doing a write access of this I/O location, the high bit of the value written, the USART Register Select(URSEL) bit, controls which one of the two registers that will be written. If URSEL is zero during a write operation,the UBRRH value will be updated. If URSEL is one, the UCSRC setting will be updated.

The following code examples show how to access the two registers.


As the code examples illustrate, write accesses of the two registers are relatively unaffected of the sharing of I/Olocation.

20.10.2 Read AccessDoing a read access to the UBRRH or the UCSRC Register is a more complex operation. However, in most appli-cations, it is rarely necessary to read any of these registers.

The read access is controlled by a timed sequence. Reading the I/O location once returns the UBRRH Registercontents. If the register location was read in previous system clock cycle, reading the register in the current clockcycle will return the UCSRC contents. Note that the timed sequence for reading the UCSRC is an atomic operation.Interrupts must therefore be controlled (for example by disabling interrupts globally) during the read operation.


:.

; Set UBRRH to 2

ldi r16,0x02

out UBRRH,r16

:.

; Set the USBS and the UCSZ1 bit to one, and

; the remaining bits to zero.

ldi r16,(1<<URSEL)|(1<<USBS)|(1<<UCSZ1)

out UCSRC,r16

:.

C Code Example(1)

:.

/* Set UBRRH to 2 */

UBRRH = 0x02;

:.

/* Set the USBS and the UCSZ1 bit to one, and */

/* the remaining bits to zero. */

UCSRC = (1<<URSEL)|(1<<USBS)|(1<<UCSZ1);

:.


The following code example shows how to read the UCSRC Register contents.


The assembly code example returns the UCSRC value in r16.

Reading the UBRRH contents is not an atomic operation and therefore it can be read as an ordinary register, aslong as the previous instruction did not access the register location.


USART_ReadUCSRC:

; Read UCSRC

in r16,UBRRH

in r16,UCSRC

ret

C Code Example(1)

unsigned char USART_ReadUCSRC( void )

{

unsigned char ucsrc;

/* Read UCSRC */

ucsrc = UBRRH;

ucsrc = UCSRC;

return ucsrc;

}



20.11.1 UDR – USART I/O Data Register

The USART Transmit Data Buffer Register and USART Receive Data Buffer Registers share the same I/O addressreferred to as USART Data Register or UDR. The Transmit Data Buffer Register (TXB) will be the destination fordata written to the UDR Register location. Reading the UDR Register location will return the contents of theReceive Data Buffer Register (RXB).

For 5-bit, 6-bit, or 7-bit characters the upper unused bits will be ignored by the Transmitter and set to zero by theReceiver.

The transmit buffer can only be written when the UDRE Flag in the UCSRA Register is set. Data written to UDRwhen the UDRE Flag is not set, will be ignored by the USART Transmitter. When data is written to the transmit buf-fer, and the Transmitter is enabled, the Transmitter will load the data into the transmit Shift Register when the ShiftRegister is empty. Then the data will be serially transmitted on the TxD pin.

The receive buffer consists of a two level FIFO. The FIFO will change its state whenever the receive buffer isaccessed. Due to this behavior of the receive buffer, do not use read modify write instructions (SBI and CBI) on thislocation. Be careful when using bit test instructions (SBIC and SBIS), since these also will change the state of theFIFO.

20.11.2 UCSRA – USART Control and Status Register A

• Bit 7 – RXC: USART Receive CompleteThis flag bit is set when there are unread data in the receive buffer and cleared when the receive buffer is empty(that is, does not contain any unread data). If the receiver is disabled, the receive buffer will be flushed and conse-quently the RXC bit will become zero. The RXC Flag can be used to generate a Receive Complete interrupt (seedescription of the RXCIE bit).

• Bit 6 – TXC: USART Transmit CompleteThis flag bit is set when the entire frame in the transmit Shift Register has been shifted out and there are no newdata currently present in the transmit buffer (UDR). The TXC Flag bit is automatically cleared when a transmit com-plete interrupt is executed, or it can be cleared by writing a one to its bit location. The TXC Flag can generate aTransmit Complete interrupt (see description of the TXCIE bit).

• Bit 5 – UDRE: USART Data Register EmptyThe UDRE Flag indicates if the transmit buffer (UDR) is ready to receive new data. If UDRE is one, the buffer isempty, and therefore ready to be written. The UDRE Flag can generate a Data Register empty Interrupt (seedescription of the UDRIE bit).

UDRE is set after a reset to indicate that the transmitter is ready.

• Bit 4 – FE: Frame Error

Bit 7 6 5 4 3 2 1 0

RXB[7:0] UDR (Read)

TXB[7:0] UDR (Write)



Bit 7 6 5 4 3 2 1 0

RXC TXC UDRE FE DOR PE U2X MPCM UCSRA

Read/Write R R/W R R R R R/W R/W



This bit is set if the next character in the receive buffer had a Frame Error when received, that is, when the first stopbit of the next character in the receive buffer is zero. This bit is valid until the receive buffer (UDR) is read. The FEbit is zero when the stop bit of received data is one. Always set this bit to zero when writing to UCSRA.

• Bit 3 – DOR: Data OverRunThis bit is set if a Data OverRun condition is detected. A Data OverRun occurs when the receive buffer is full (twocharacters), it is a new character waiting in the receive Shift Register, and a new start bit is detected. This bit isvalid until the receive buffer (UDR) is read. Always set this bit to zero when writing to UCSRA.

• Bit 2 – PE: Parity ErrorThis bit is set if the next character in the receive buffer had a Parity Error when received and the parity checkingwas enabled at that point (UPM1 = 1). This bit is valid until the receive buffer (UDR) is read. Always set this bit tozero when writing to UCSRA.

• Bit 1 – U2X: Double the USART Transmission SpeedThis bit only has effect for the asynchronous operation. Write this bit to zero when using synchronous operation.

Writing this bit to one will reduce the divisor of the baud rate divider from 16 to 8 effectively doubling the transferrate for asynchronous communication.

• Bit 0 – MPCM: Multi-processor Communication ModeThis bit enables the Multi-processor Communication mode. When the MPCM bit is written to one, all the incomingframes received by the USART receiver that do not contain address information will be ignored. The transmitter isunaffected by the MPCM setting. For more detailed information see “Multi-processor Communication Mode” onpage 164.

20.11.3 UCSRB – USART Control and Status Register B

• Bit 7 – RXCIE: RX Complete Interrupt EnableWriting this bit to one enables interrupt on the RXC Flag. A USART Receive Complete Interrupt will be generatedonly if the RXCIE bit is written to one, the Global Interrupt Flag in SREG is written to one and the RXC bit inUCSRA is set.

• Bit 6 – TXCIE: TX Complete Interrupt EnableWriting this bit to one enables interrupt on the TXC Flag. A USART Transmit Complete Interrupt will be generatedonly if the TXCIE bit is written to one, the Global Interrupt Flag in SREG is written to one and the TXC bit in UCSRAis set.

• Bit 5 – UDRIE: USART Data Register Empty Interrupt EnableWriting this bit to one enables interrupt on the UDRE Flag. A Data Register Empty Interrupt will be generated onlyif the UDRIE bit is written to one, the Global Interrupt Flag in SREG is written to one and the UDRE bit in UCSRA isset.

• Bit 4 – RXEN: Receiver EnableWriting this bit to one enables the USART Receiver. The Receiver will override normal port operation for the RxDpin when enabled. Disabling the Receiver will flush the receive buffer invalidating the FE, DOR, and PE Flags.

Bit 7 6 5 4 3 2 1 0

RXCIE TXCIE UDRIE RXEN TXEN UCSZ2 RXB8 TXB8 UCSRB

Read/Write R/W R/W R/W R/W R/W R/W R R/W



• Bit 3 – TXEN: Transmitter EnableWriting this bit to one enables the USART Transmitter. The Transmitter will override normal port operation for theTxD pin when enabled. The disabling of the Transmitter (writing TXEN to zero) will not become effective until ongo-ing and pending transmissions are completed, that is, when the transmit Shift Register and transmit Buffer Registerdo not contain data to be transmitted. When disabled, the transmitter will no longer override the TxD port.

• Bit 2 – UCSZ2: Character SizeThe UCSZ2 bits combined with the UCSZ1:0 bit in UCSRC sets the number of data bits (Character Size) in a framethe receiver and transmitter use.

• Bit 1 – RXB8: Receive Data Bit 8RXB8 is the ninth data bit of the received character when operating with serial frames with nine data bits. Must beread before reading the low bits from UDR.

• Bit 0 – TXB8: Transmit Data Bit 8TXB8 is the ninth data bit in the character to be transmitted when operating with serial frames with nine data bits.Must be written before writing the low bits to UDR.

20.11.4 UCSRC – USART Control and Status Register C

The UCSRC Register shares the same I/O location as the UBRRH Register. See the “Accessing UBRRH/ UCSRCRegisters” on page 165 section which describes how to access this register.

• Bit 7 – URSEL: Register SelectThis bit selects between accessing the UCSRC or the UBRRH Register. It is read as one when reading UCSRC.The URSEL must be one when writing the UCSRC.

• Bit 6 – UMSEL: USART Mode SelectThis bit selects between Asynchronous and Synchronous mode of operation.

• Bit 5:4 – UPM1:0: Parity ModeThese bits enable and set type of parity generation and check. If enabled, the transmitter will automatically gener-ate and send the parity of the transmitted data bits within each frame. The Receiver will generate a parity value forthe incoming data and compare it to the UPM0 setting. If a mismatch is detected, the PE Flag in UCSRA will be set.

Bit 7 6 5 4 3 2 1 0

URSEL UMSEL UPM1 UPM0 USBS UCSZ1 UCSZ0 UCPOL UCSRC



Table 20-4. UMSEL Bit Settings

UMSEL Mode

0 Asynchronous Operation

1 Synchronous Operation


• Bit 3 – USBS: Stop Bit SelectThis bit selects the number of Stop Bits to be inserted by the Transmitter. The Receiver ignores this setting.

• Bit 2:1 – UCSZ1:0: Character SizeThe UCSZ1:0 bits combined with the UCSZ2 bit in UCSRB sets the number of data bits (Character Size) in a framethe Receiver and Transmitter use.

• Bit 0 – UCPOL: Clock PolarityThis bit is used for Synchronous mode only. Write this bit to zero when Asynchronous mode is used. The UCPOLbit sets the relationship between data output change and data input sample, and the synchronous clock (XCK).

Table 20-5. UPM Bits Settings

UPM1 UPM0 Parity Mode

0 0 Disabled

0 1 Reserved

1 0 Enabled, Even Parity

1 1 Enabled, Odd Parity

Table 20-6. USBS Bit Settings

USBS Stop Bit(s)

0 1-bit

1 2-bit

Table 20-7. UCSZ Bits Settings

UCSZ2 UCSZ1 UCSZ0 Character Size

0 0 0 5-bit

0 0 1 6-bit

0 1 0 7-bit

0 1 1 8-bit

1 0 0 Reserved

1 0 1 Reserved

1 1 0 Reserved

1 1 1 9-bit

Table 20-8. UCPOL Bit Settings

UCPOLTransmitted Data Changed (Output of TxD Pin)

Received Data Sampled (Input on RxD Pin)

0 Rising XCK Edge Falling XCK Edge

1 Falling XCK Edge Rising XCK Edge


20.11.5 UBRRL and UBRRH – USART Baud Rate Registers

The UBRRH Register shares the same I/O location as the UCSRC Register. See the “Accessing UBRRH/ UCSRCRegisters” on page 165 section which describes how to access this register.

• Bit 15 – URSEL: Register SelectThis bit selects between accessing the UBRRH or the UCSRC Register. It is read as zero when reading UBRRH.The URSEL must be zero when writing the UBRRH.

• Bit 14:12 – Reserved BitsThese bits are reserved for future use. For compatibility with future devices, these bit must be written to zero whenUBRRH is written.

• Bit 11:0 – UBRR11:0: USART Baud Rate RegisterThis is a 12-bit register which contains the USART baud rate. The UBRRH contains the four most significant bits,and the UBRRL contains the 8 least significant bits of the USART baud rate. Ongoing transmissions by the trans-mitter and receiver will be corrupted if the baud rate is changed. Writing UBRRL will trigger an immediate update ofthe baud rate prescaler.

20.12 Examples of Baud Rate SettingFor standard crystal and resonator frequencies, the most commonly used baud rates for asynchronous operationcan be generated by using the UBRR settings in Table 19-9. UBRR values which yield an actual baud rate differingless than 0.5% from the target baud rate, are bold in the table. Higher error ratings are acceptable, but the receiverwill have less noise resistance when the error ratings are high, especially for large serial frames (see “Asynchro-nous Operational Range” on page 163). The error values are calculated using the following equation:

Bit 15 14 13 12 11 10 9 8

URSEL – – – UBRR[11:8] UBRRH

UBRR[7:0] UBRRL

7 6 5 4 3 2 1 0

Read/Write R/W R R R R/W R/W R/W R/W



0 0 0 0 0 0 0 0

Error[%]BaudRateClosest Match

BaudRate-------------------------------------------------------- 1– 100%=

Table 20-9. Examples of UBRR Settings for Commonly Used Oscillator Frequencies

Baud Rate (bps)

fosc = 1.0000MHz fosc = 1.8432MHz fosc = 2.0000MHz

U2X = 0 U2X = 1 U2X = 0 U2X = 1 U2X = 0 U2X = 1

UBRR Error UBRR Error UBRR Error UBRR Error UBRR Error UBRR Error

2400 25 0.2% 51 0.2% 47 0.0% 95 0.0% 51 0.2% 103 0.2%

4800 12 0.2% 25 0.2% 23 0.0% 47 0.0% 25 0.2% 51 0.2%

9600 6 -7.0% 12 0.2% 11 0.0% 23 0.0% 12 0.2% 25 0.2%

14.4k 3 8.5% 8 -3.5% 7 0.0% 15 0.0% 8 -3.5% 16 2.1%


19.2k 2 8.5% 6 -7.0% 5 0.0% 11 0.0% 6 -7.0% 12 0.2%

28.8k 1 8.5% 3 8.5% 3 0.0% 7 0.0% 3 8.5% 8 -3.5%

38.4k 1 -18.6% 2 8.5% 2 0.0% 5 0.0% 2 8.5% 6 -7.0%

57.6k 0 8.5% 1 8.5% 1 0.0% 3 0.0% 1 8.5% 3 8.5%

76.8k – – 1 -18.6% 1 -25.0% 2 0.0% 1 -18.6% 2 8.5%

115.2k – – 0 8.5% 0 0.0% 1 0.0% 0 8.5% 1 8.5%

230.4k – – – – – – 0 0.0% – – – –

250k – – – – – – – – – – 0 0.0%

Max (1) 62.5Kbps 125Kbps 115.2Kbps 230.4Kbps 125Kbps 250Kbps

1. UBRR = 0, Error = 0.0%

Table 20-9. Examples of UBRR Settings for Commonly Used Oscillator Frequencies (Continued)

Baud Rate (bps)


U2X = 0 U2X = 1 U2X = 0 U2X = 1 U2X = 0 U2X = 1




Baud Rate (bps)


U2X = 0 U2X = 1 U2X = 0 U2X = 1 U2X = 0 U2X = 1


2400 95 0.0% 191 0.0% 103 0.2% 207 0.2% 191 0.0% 383 0.0%

4800 47 0.0% 95 0.0% 51 0.2% 103 0.2% 95 0.0% 191 0.0%

9600 23 0.0% 47 0.0% 25 0.2% 51 0.2% 47 0.0% 95 0.0%

14.4k 15 0.0% 31 0.0% 16 2.1% 34 -0.8% 31 0.0% 63 0.0%

19.2k 11 0.0% 23 0.0% 12 0.2% 25 0.2% 23 0.0% 47 0.0%

28.8k 7 0.0% 15 0.0% 8 -3.5% 16 2.1% 15 0.0% 31 0.0%

38.4k 5 0.0% 11 0.0% 6 -7.0% 12 0.2% 11 0.0% 23 0.0%

57.6k 3 0.0% 7 0.0% 3 8.5% 8 -3.5% 7 0.0% 15 0.0%

76.8k 2 0.0% 5 0.0% 2 8.5% 6 -7.0% 5 0.0% 11 0.0%

115.2k 1 0.0% 3 0.0% 1 8.5% 3 8.5% 3 0.0% 7 0.0%

230.4k 0 0.0% 1 0.0% 0 8.5% 1 8.5% 1 0.0% 3 0.0%

250k 0 -7.8% 1 -7.8% 0 0.0% 1 0.0% 1 -7.8% 3 -7.8%

0.5M – – 0 -7.8% – – 0 0.0% 0 -7.8% 1 -7.8%

1M – – – – – – – – – – 0 -7.8%

Max (1) 230.4Kbps 460.8Kbps 250Kbps 0.5Mbps 460.8Kbps 921.6Kbps

1. UBRR = 0, Error = 0.0%



Baud Rate (bps)


U2X = 0 U2X = 1 U2X = 0 U2X = 1 U2X = 0 U2X = 1


2400 207 0.2% 416 -0.1% 287 0.0% 575 0.0% 383 0.0% 767 0.0%

4800 103 0.2% 207 0.2% 143 0.0% 287 0.0% 191 0.0% 383 0.0%

9600 51 0.2% 103 0.2% 71 0.0% 143 0.0% 95 0.0% 191 0.0%

14.4k 34 -0.8% 68 0.6% 47 0.0% 95 0.0% 63 0.0% 127 0.0%

19.2k 25 0.2% 51 0.2% 35 0.0% 71 0.0% 47 0.0% 95 0.0%

28.8k 16 2.1% 34 -0.8% 23 0.0% 47 0.0% 31 0.0% 63 0.0%

38.4k 12 0.2% 25 0.2% 17 0.0% 35 0.0% 23 0.0% 47 0.0%

57.6k 8 -3.5% 16 2.1% 11 0.0% 23 0.0% 15 0.0% 31 0.0%

76.8k 6 -7.0% 12 0.2% 8 0.0% 17 0.0% 11 0.0% 23 0.0%

115.2k 3 8.5% 8 -3.5% 5 0.0% 11 0.0% 7 0.0% 15 0.0%

230.4k 1 8.5% 3 8.5% 2 0.0% 5 0.0% 3 0.0% 7 0.0%

250k 1 0.0% 3 0.0% 2 -7.8% 5 -7.8% 3 -7.8% 6 5.3%

0.5M 0 0.0% 1 0.0% – – 2 -7.8% 1 -7.8% 3 -7.8%

1M – – 0 0.0% – – – – 0 -7.8% 1 -7.8%

Max (1) 0.5Mbps 1Mbps 691.2Kbps 1.3824Mbps 921.6Kbps 1.8432Mbps

1. UBRR = 0, Error = 0.0%



Baud Rate (bps)

fosc = 16.0000MHz

U2X = 0 U2X = 1

UBRR Error UBRR Error

2400 416 -0.1% 832 0.0%

4800 207 0.2% 416 -0.1%

9600 103 0.2% 207 0.2%

14.4k 68 0.6% 138 -0.1%

19.2k 51 0.2% 103 0.2%

28.8k 34 -0.8% 68 0.6%

38.4k 25 0.2% 51 0.2%

57.6k 16 2.1% 34 -0.8%

76.8k 12 0.2% 25 0.2%

115.2k 8 -3.5% 16 2.1%

230.4k 3 8.5% 8 -3.5%

250k 3 0.0% 7 0.0%

0.5M 1 0.0% 3 0.0%

1M 0 0.0% 1 0.0%

Max (1)

1. UBRR = 0, Error = 0.0%

1Mbps 2Mbps


21. Two-wire Serial Interface

21.1 Features• Simple Yet Powerful and Flexible Communication Interface, Only Two Bus Lines Needed• Both Master and Slave Operation Supported• Device Can Operate as Transmitter or Receiver• 7-bit Address Space allows up to 128 Different Slave Addresses• Multi-master Arbitration Support• Up to 400kHz Data Transfer Speed• Slew-rate Limited Output Drivers• Noise Suppression Circuitry Rejects Spikes on Bus Lines• Fully Programmable Slave Address with General Call Support• Address Recognition causes Wake-up when AVR is in Sleep Mode

21.2 Two-wire Serial Interface Bus DefinitionThe Two-wire Serial Interface (TWI) is ideally suited for typical microcontroller applications. The TWI protocolallows the systems designer to interconnect up to 128 different devices using only two bi-directional bus lines, onefor clock (SCL) and one for data (SDA). The only external hardware needed to implement the bus is a single pull-up resistor for each of the TWI bus lines. All devices connected to the bus have individual addresses, and mecha-nisms for resolving bus contention are inherent in the TWI protocol.

Figure 21-1. TWI Bus Interconnection

Device 1 Device 2 Device 3 Device n

SDA

SCL

........ R1 R2

VCC


21.2.1 TWI TerminologyThe following definitions are frequently encountered in this section.

21.2.2 Electrical InterconnectionAs depicted in Figure 20-1, both bus lines are connected to the positive supply voltage through pull-up resistors.The bus drivers of all TWI-compliant devices are open-drain or open-collector. This implements a wired-AND func-tion which is essential to the operation of the interface. A low level on a TWI bus line is generated when one ormore TWI devices output a zero. A high level is output when all TWI devices tri-state their outputs, allowing thepull-up resistors to pull the line high. Note that all AVR devices connected to the TWI bus must be powered in orderto allow any bus operation.

The number of devices that can be connected to the bus is only limited by the bus capacitance limit of 400 pF andthe 7-bit slave address space. A detailed specification of the electrical characteristics of the TWI is given in “Two-wire Serial Interface Characteristics” on page 300. Two different sets of specifications are presented there, one rel-evant for bus speeds below 100kHz, and one valid for bus speeds up to 400kHz.

21.3 Data Transfer and Frame Format

21.3.1 Transferring BitsEach data bit transferred on the TWI bus is accompanied by a pulse on the clock line. The level of the data linemust be stable when the clock line is high. The only exception to this rule is for generating start and stopconditions.

Figure 21-2. Data Validity

21.3.2 START and STOP ConditionsThe master initiates and terminates a data transmission. The transmission is initiated when the master issues aSTART condition on the bus, and it is terminated when the master issues a STOP condition. Between a STARTand a STOP condition, the bus is considered busy, and no other master should try to seize control of the bus. Aspecial case occurs when a new START condition is issued between a START and STOP condition. This isreferred to as a REPEATED START condition, and is used when the master wishes to initiate a new transfer with-out releasing control of the bus. After a REPEATED START, the bus is considered busy until the next STOP. Thisis identical to the START behavior, and therefore START is used to describe both START and REPEATED START

Table 21-1. TWI Terminology

Term Description

Master The device that initiates and terminates a transmission. The master also generates the SCL clock.

Slave The device addressed by a master.

Transmitter The device placing data on the bus.

Receiver The device reading data from the bus.

SDA

SCL

Data Stable Data Stable

Data Change


for the remainder of this datasheet, unless otherwise noted. As depicted below, START and STOP conditions aresignalled by changing the level of the SDA line when the SCL line is high.

Figure 21-3. START, REPEATED START, and STOP Conditions

21.3.3 Address Packet FormatAll address packets transmitted on the TWI bus are nine bits long, consisting of seven address bits, oneREAD/WRITE control bit and an acknowledge bit. If the READ/WRITE bit is set, a read operation is to be per-formed, otherwise a write operation should be performed. When a slave recognizes that it is being addressed, itshould acknowledge by pulling SDA low in the ninth SCL (ACK) cycle. If the addressed slave is busy, or for someother reason can not service the master’s request, the SDA line should be left high in the ACK clock cycle. Themaster can then transmit a STOP condition, or a REPEATED START condition to initiate a new transmission. Anaddress packet consisting of a slave address and a READ or a WRITE bit is called SLA+R or SLA+W, respectively.

The MSB of the address byte is transmitted first. Slave addresses can freely be allocated by the designer, but theaddress 0000 000 is reserved for a general call.

When a general call is issued, all slaves should respond by pulling the SDA line low in the ACK cycle. A generalcall is used when a master wishes to transmit the same message to several slaves in the system. When the gen-eral call address followed by a Write bit is transmitted on the bus, all slaves set up to acknowledge the general callwill pull the SDA line low in the ack cycle. The following data packets will then be received by all the slaves thatacknowledged the general call. Note that transmitting the general call address followed by a Read bit is meaning-less, as this would cause contention if several slaves started transmitting different data.

All addresses of the format 1111 xxx should be reserved for future purposes.

Figure 21-4. Address Packet Format

21.3.4 Data Packet FormatAll data packets transmitted on the TWI bus are nine bits long, consisting of one data byte and an acknowledge bit.During a data transfer, the master generates the clock and the START and STOP conditions, while the receiver isresponsible for acknowledging the reception. An Acknowledge (ACK) is signalled by the receiver pulling the SDAline low during the ninth SCL cycle. If the receiver leaves the SDA line high, a NACK is signalled. When thereceiver has received the last byte, or for some reason cannot receive any more bytes, it should inform the trans-mitter by sending a NACK after the final byte. The MSB of the data byte is transmitted first.

SDA

SCL

START STOPREPEATED STARTSTOP START

SDA

SCL

START

1 2 7 8 9

Addr MSB Addr LSB R/W ACK


Figure 21-5. Data Packet Format

21.3.5 Combining Address and Data Packets into a TransmissionA transmission basically consists of a START condition, a SLA+R/W, one or more data packets and a STOP con-dition. An empty message, consisting of a START followed by a STOP condition, is illegal. Note that the wired-ANDing of the SCL line can be used to implement handshaking between the master and the slave. The slave canextend the SCL low period by pulling the SCL line low. This is useful if the clock speed set up by the master is toofast for the slave, or the slave needs extra time for processing between the data transmissions. The slave extend-ing the SCL low period will not affect the SCL high period, which is determined by the master. As a consequence,the slave can reduce the TWI data transfer speed by prolonging the SCL duty cycle.

Figure 20-6 shows a typical data transmission. Note that several data bytes can be transmitted between theSLA+R/W and the STOP condition, depending on the software protocol implemented by the application software.

Figure 21-6. Typical Data Transmission

21.4 Multi-master Bus Systems, Arbitration and SynchronizationThe TWI protocol allows bus systems with several masters. Special concerns have been taken in order to ensurethat transmissions will proceed as normal, even if two or more masters initiate a transmission at the same time.Two problems arise in multi-master systems:

• An algorithm must be implemented allowing only one of the masters to complete the transmission. All other masters should cease transmission when they discover that they have lost the selection process. This selection process is called arbitration. When a contending master discovers that it has lost the arbitration process, it should immediately switch to slave mode to check whether it is being addressed by the winning master. The fact that multiple masters have started transmission at the same time should not be detectable to the slaves, that is, the data being transferred on the bus must not be corrupted.

• Different masters may use different SCL frequencies. A scheme must be devised to synchronize the serial clocks from all masters, in order to let the transmission proceed in a lockstep fashion. This will facilitate the arbitration process.

The wired-ANDing of the bus lines is used to solve both these problems. The serial clocks from all masters will bewired-ANDed, yielding a combined clock with a high period equal to the one from the master with the shortest highperiod. The low period of the combined clock is equal to the low period of the master with the longest low period.

1 2 7 8 9

Data MSB Data LSB ACK

AggregateSDA

SDA fromTransmitter

SDA fromreceiverR

SCL fromMaster

SLA+R/W Data ByteSTOP, REPEATED

START or NextData Byte

1 2 7 8 9

Data Byte

Data MSB Data LSB ACK

SDA

SCL

START

1 2 7 8 9

Addr MSB Addr LSB R/W ACK

SLA+R/W STOP


Note that all masters listen to the SCL line, effectively starting to count their SCL high and low time-out periodswhen the combined SCL line goes high or low, respectively.

Figure 21-7. SCL Synchronization between Multiple Masters

Arbitration is carried out by all masters continuously monitoring the SDA line after outputting data. If the value readfrom the SDA line does not match the value the master had output, it has lost the arbitration. Note that a mastercan only lose arbitration when it outputs a high SDA value while another master outputs a low value. The losingmaster should immediately go to slave mode, checking if it is being addressed by the winning master. The SDAline should be left high, but losing masters are allowed to generate a clock signal until the end of the current data oraddress packet. Arbitration will continue until only one master remains, and this may take many bits. If severalmasters are trying to address the same slave, arbitration will continue into the data packet.

Figure 21-8. Arbitration between Two Masters

Note that arbitration is not allowed between:

• A REPEATED START condition and a data bit

• A STOP condition and a data bit

TA low TA high

SCL fromMaster A

SCL fromMaster B

SCL busLine

TBlow TBhigh

Masters StartCounting Low Period

Masters StartCounting High Period

SDA fromMaster A

SDA fromMaster B

SDA Line

SynchronizedSCL Line

START Master A LosesArbitration, SDAA SDA


• A REPEATED START and a STOP condition

It is the user software’s responsibility to ensure that these illegal arbitration conditions never occur. This impliesthat in multi-master systems, all data transfers must use the same composition of SLA+R/W and data packets. Inother words: All transmissions must contain the same number of data packets, otherwise the result of the arbitra-tion is undefined.

21.5 Overview of the TWI ModuleThe TWI module is comprised of several submodules, as shown in Figure 20-9. All registers drawn in a thick lineare accessible through the AVR data bus.

Figure 21-9. Overview of the TWI Module

21.5.1 SCL and SDA PinsThese pins interface the AVR TWI with the rest of the MCU system. The output drivers contain a slew-rate limiter inorder to conform to the TWI specification. The input stages contain a spike suppression unit removing spikesshorter than 50 ns. Note that the internal pullups in the AVR pads can be enabled by setting the PORT bits corre-sponding to the SCL and SDA pins, as explained in the I/O Port section. The internal pull-ups can in some systemseliminate the need for external ones.

21.5.2 Bit Rate Generator UnitThis unit controls the period of SCL when operating in a Master mode. The SCL period is controlled by settings inthe TWI Bit Rate Register (TWBR) and the Prescaler bits in the TWI Status Register (TWSR). Slave operationdoes not depend on Bit Rate or Prescaler settings, but the CPU clock frequency in the slave must be at least 16times higher than the SCL frequency. Note that slaves may prolong the SCL low period, thereby reducing the aver-age TWI bus clock period. The SCL frequency is generated according to the following equation:

TWI Unit

Address Register(TWAR)

Address Match Unit

Address Comparator

Control Unit

Control Register(TWCR)

Status Register(TWSR)

State Machine andStatus control

SCL

Slew-rateControl

SpikeFilter

SDA

Slew-rateControl

SpikeFilter

Bit Rate Generator

Bit Rate Register(TWBR)

Prescaler

Bus Interface Unit

START / STOPControl

Arbitration detection Ack

Spike Suppression

Address/Data ShiftRegister (TWDR)

SCL frequency CPU Clock frequency

16 2(TWBR) 4TWPS+

-----------------------------------------------------------=


• TWBR = Value of the TWI Bit Rate Register

• TWPS = Value of the prescaler bits in the TWI Status Register

Note: Pull-up resistor values should be selected according to the SCL frequency and the capacitive bus line load. See Table 27-2 on page 300 for value of pull-up resistor.

21.5.3 Bus Interface UnitThis unit contains the Data and Address Shift Register (TWDR), a START/STOP Controller and Arbitration detec-tion hardware. The TWDR contains the address or data bytes to be transmitted, or the address or data bytesreceived. In addition to the 8-bit TWDR, the Bus Interface Unit also contains a register containing the (N)ACK bit tobe transmitted or received. This (N)ACK Register is not directly accessible by the application software. However,when receiving, it can be set or cleared by manipulating the TWI Control Register (TWCR). When in Transmittermode, the value of the received (N)ACK bit can be determined by the value in the TWSR.

The START/STOP Controller is responsible for generation and detection of START, REPEATED START, andSTOP conditions. The START/STOP controller is able to detect START and STOP conditions even when the AVRMCU is in one of the sleep modes, enabling the MCU to wake up if addressed by a master.

If the TWI has initiated a transmission as master, the Arbitration Detection hardware continuously monitors thetransmission trying to determine if arbitration is in process. If the TWI has lost an arbitration, the Control Unit isinformed. Correct action can then be taken and appropriate status codes generated.

21.5.4 Address Match UnitThe Address Match unit checks if received address bytes match the 7-bit address in the TWI Address Register(TWAR). If the TWI General Call Recognition Enable (TWGCE) bit in the TWAR is written to one, all incomingaddress bits will also be compared against the General Call address. Upon an address match, the Control Unit isinformed, allowing correct action to be taken. The TWI may or may not acknowledge its address, depending on set-tings in the TWCR. The Address Match unit is able to compare addresses even when the AVR MCU is in sleepmode, enabling the MCU to wake up if addressed by a master.

21.5.5 Control UnitThe Control unit monitors the TWI bus and generates responses corresponding to settings in the TWI Control Reg-ister (TWCR). When an event requiring the attention of the application occurs on the TWI bus, the TWI InterruptFlag (TWINT) is asserted. In the next clock cycle, the TWI Status Register (TWSR) is updated with a status codeidentifying the event. The TWSR only contains relevant status information when the TWI Interrupt Flag is asserted.At all other times, the TWSR contains a special status code indicating that no relevant status information is avail-able. As long as the TWINT Flag is set, the SCL line is held low. This allows the application software to complete itstasks before allowing the TWI transmission to continue.

The TWINT Flag is set in the following situations:

• After the TWI has transmitted a START/REPEATED START condition

• After the TWI has transmitted SLA+R/W

• After the TWI has transmitted an address byte

• After the TWI has lost arbitration

• After the TWI has been addressed by own slave address or general call

• After the TWI has received a data byte

• After a STOP or REPEATED START has been received while still addressed as a slave

• When a bus error has occurred due to an illegal START or STOP condition


21.6 Using the TWIThe AVR TWI is byte-oriented and interrupt based. Interrupts are issued after all bus events, like reception of abyte or transmission of a START condition. Because the TWI is interrupt-based, the application software is free tocarry on other operations during a TWI byte transfer. Note that the TWI Interrupt Enable (TWIE) bit in TWCRtogether with the Global Interrupt Enable bit in SREG allow the application to decide whether or not assertion of theTWINT Flag should generate an interrupt request. If the TWIE bit is cleared, the application must poll the TWINTFlag in order to detect actions on the TWI bus.

When the TWINT Flag is asserted, the TWI has finished an operation and awaits application response. In thiscase, the TWI Status Register (TWSR) contains a value indicating the current state of the TWI bus. The applicationsoftware can then decide how the TWI should behave in the next TWI bus cycle by manipulating the TWCR andTWDR Registers.

Figure 20-10 is a simple example of how the application can interface to the TWI hardware. In this example, a mas-ter wishes to transmit a single data byte to a slave. This description is quite abstract, a more detailed explanationfollows later in this section. A simple code example implementing the desired behavior is also presented.

Figure 21-10. Interfacing the Application to the TWI in a Typical Transmission

1. The first step in a TWI transmission is to transmit a START condition. This is done by writing a specific value into TWCR, instructing the TWI hardware to transmit a START condition. Which value to write is described later on. However, it is important that the TWINT bit is set in the value written. Writing a one to TWINT clears the Flag. The TWI will not start any operation as long as the TWINT bit in TWCR is set. Immediately after the application has cleared TWINT, the TWI will initiate transmission of the START condition.

2. When the START condition has been transmitted, the TWINT Flag in TWCR is set, and TWSR is updated with a status code indicating that the START condition has successfully been sent.

3. The application software should now examine the value of TWSR, to make sure that the START condition was successfully transmitted. If TWSR indicates otherwise, the application software might take some spe-cial action, like calling an error routine. Assuming that the status code is as expected, the application must load SLA+W into TWDR. Remember that TWDR is used both for address and data. After TWDR has been loaded with the desired SLA+W, a specific value must be written to TWCR, instructing the TWI hardware to transmit the SLA+W present in TWDR. Which value to write is described later on. However, it is impor-tant that the TWINT bit is set in the value written. Writing a one to TWINT clears the flag. The TWI will not start any operation as long as the TWINT bit in TWCR is set. Immediately after the application has cleared TWINT, the TWI will initiate transmission of the address packet.

START SLA+W A Data A STOP

1. Application writes to TWCR

to initiate transmission of

START

2. TWINT set.Status code indicatesSTART condition sent

4. TWINT set.Status code indicates

SLA+W sent, ACKreceived

6. TWINT set.Status code indicates

data sent, ACK received

5. Check TWSR to see if SLA+W wassent and ACK received.

Application loads data into TWDR, andloads appropriate control signals intoTWCR, making sure that TWINT is

written to one

7. Check TWSR to see if data was sentand ACK received.

Application loads appropriate controlsignals to send STOP into TWCR,

making sure that TWINT is written to one

TWI bus

IndicatesTWINT set

ApplicationAction

TWIHardware

Action

3. Check TWSR to see if START was sendt. Application loads SLA+W into TWDR, and loads appropriate control signals into TWCR, making sure that TWINT is written to one, and TWSTA

is written to zero


4. When the address packet has been transmitted, the TWINT Flag in TWCR is set, and TWSR is updated with a status code indicating that the address packet has successfully been sent. The status code will also reflect whether a slave acknowledged the packet or not.

5. The application software should now examine the value of TWSR, to make sure that the address packet was successfully transmitted, and that the value of the ACK bit was as expected. If TWSR indicates other-wise, the application software might take some special action, like calling an error routine. Assuming that the status code is as expected, the application must load a data packet into TWDR. Subsequently, a spe-cific value must be written to TWCR, instructing the TWI hardware to transmit the data packet present in TWDR. Which value to write is described later on. However, it is important that the TWINT bit is set in the value written. Writing a one to TWINT clears the flag. The TWI will not start any operation as long as the TWINT bit in TWCR is set. Immediately after the application has cleared TWINT, the TWI will initiate trans-mission of the data packet.

6. When the data packet has been transmitted, the TWINT Flag in TWCR is set, and TWSR is updated with a status code indicating that the data packet has successfully been sent. The status code will also reflect whether a slave acknowledged the packet or not.

7. The application software should now examine the value of TWSR, to make sure that the data packet was successfully transmitted, and that the value of the ACK bit was as expected. If TWSR indicates otherwise, the application software might take some special action, like calling an error routine. Assuming that the status code is as expected, the application must write a specific value to TWCR, instructing the TWI hard-ware to transmit a STOP condition. Which value to write is described later on. However, it is important that the TWINT bit is set in the value written. Writing a one to TWINT clears the flag. The TWI will not start any operation as long as the TWINT bit in TWCR is set. Immediately after the application has cleared TWINT, the TWI will initiate transmission of the STOP condition. Note that TWINT is NOT set after a STOP condi-tion has been sent.

Even though this example is simple, it shows the principles involved in all TWI transmissions. These can be sum-marized as follows:

• When the TWI has finished an operation and expects application response, the TWINT Flag is set. The SCL line is pulled low until TWINT is cleared.

• When the TWINT Flag is set, the user must update all TWI Registers with the value relevant for the next TWI bus cycle. As an example, TWDR must be loaded with the value to be transmitted in the next bus cycle.

• After all TWI Register updates and other pending application software tasks have been completed, TWCR is written. When writing TWCR, the TWINT bit should be set. Writing a one to TWINT clears the flag. The TWI will then commence executing whatever operation was specified by the TWCR setting.

In the following an assembly and C implementation of the example is given. Note that the code below assumes thatseveral definitions have been made, for example by using include-files.

Assembly code example C example Comments

1 ldi r16, (1<<TWINT)|(1<<TWSTA)|

(1<<TWEN)

out TWCR, r16

TWCR = (1<<TWINT)|(1<<TWSTA)|

(1<<TWEN)Send START condition

2 wait1:

in r16,TWCR

sbrs r16,TWINT

rjmp wait1

while (!(TWCR & (1<<TWINT)))

;Wait for TWINT Flag set. This indicates that the START condition has been transmitted


21.7 Transmission ModesThe TWI can operate in one of four major modes. These are named Master Transmitter (MT), Master Receiver(MR), Slave Transmitter (ST) and Slave Receiver (SR). Several of these modes can be used in the same applica-tion. As an example, the TWI can use MT mode to write data into a TWI EEPROM, MR mode to read the data backfrom the EEPROM. If other masters are present in the system, some of these might transmit data to the TWI, andthen SR mode would be used. It is the application software that decides which modes are legal.

The following sections describe each of these modes. Possible status codes are described along with figuresdetailing data transmission in each of the modes. These figures contain the following abbreviations:

S: START condition

Rs: REPEATED START condition

R: Read bit (high level at SDA)

W: Write bit (low level at SDA)

A: Acknowledge bit (low level at SDA)

A: Not acknowledge bit (high level at SDA)

3 in r16,TWSR

andi r16, 0xF8

cpi r16, START

brne ERROR

if ((TWSR & 0xF8) != START)

ERROR();Check value of TWI Status Register. Mask prescaler bits. If status different from START go to ERROR

ldi r16, SLA_W

out TWDR, r16

ldi r16, (1<<TWINT) | (1<<TWEN)

out TWCR, r16

TWDR = SLA_W;

TWCR = (1<<TWINT) | (1<<TWEN);Load SLA_W into TWDR Register. Clear TWINT bit in TWCR to start transmission of address

4 wait2:

in r16,TWCR

sbrs r16,TWINT

rjmp wait2


;Wait for TWINT Flag set. This indicates that the SLA+W has been transmitted, and ACK/NACK has been received.

5 in r16,TWSR

andi r16, 0xF8

cpi r16, MT_SLA_ACK

brne ERROR

if ((TWSR & 0xF8) != MT_SLA_ACK)

ERROR();Check value of TWI Status Register. Mask prescaler bits. If status different from MT_SLA_ACK go to ERROR

ldi r16, DATA

out TWDR, r16

ldi r16, (1<<TWINT) | (1<<TWEN)

out TWCR, r16

TWDR = DATA;

TWCR = (1<<TWINT) | (1<<TWEN);Load DATA into TWDR Register. Clear TWINT bit in TWCR to start transmission of data

6 wait3:

in r16,TWCR

sbrs r16,TWINT

rjmp wait3


;Wait for TWINT Flag set. This indicates that the DATA has been transmitted, and ACK/NACK has been received.

7 in r16,TWSR

andi r16, 0xF8

cpi r16, MT_DATA_ACK

brne ERROR

if ((TWSR & 0xF8) != MT_DATA_ACK)

ERROR();Check value of TWI Status Register. Mask prescaler bits. If status different from MT_DATA_ACK go to ERROR

ldi r16, (1<<TWINT)|(1<<TWEN)|

(1<<TWSTO)

out TWCR, r16

TWCR = (1<<TWINT)|(1<<TWEN)|

(1<<TWSTO);Transmit STOP condition

Assembly code example C example Comments


Data: 8-bit data byte

P: STOP condition

SLA: Slave Address

In Figure 20-12 to Figure 20-18, circles are used to indicate that the TWINT Flag is set. The numbers in the circlesshow the status code held in TWSR, with the prescaler bits masked to zero. At these points, actions must be takenby the application to continue or complete the TWI transfer. The TWI transfer is suspended until the TWINT Flag iscleared by software.

When the TWINT Flag is set, the status code in TWSR is used to determine the appropriate software action. Foreach status code, the required software action and details of the following serial transfer are given in Table 20-2 toTable 20-5. Note that the prescaler bits are masked to zero in these tables.

21.7.1 Master Transmitter ModeIn the Master Transmitter mode, a number of data bytes are transmitted to a slave receiver (see Figure 20-11). Inorder to enter a Master mode, a START condition must be transmitted. The format of the following address packetdetermines whether Master Transmitter or Master Receiver mode is to be entered. If SLA+W is transmitted, MTmode is entered, if SLA+R is transmitted, MR mode is entered. All the status codes mentioned in this sectionassume that the prescaler bits are zero or are masked to zero.

Figure 21-11. Data Transfer in Master Transmitter Mode

A START condition is sent by writing the following value to TWCR:

TWEN must be set to enable the Two-wire Serial Interface, TWSTA must be written to one to transmit a STARTcondition and TWINT must be written to one to clear the TWINT Flag. The TWI will then test the Two-wire SerialBus and generate a START condition as soon as the bus becomes free. After a START condition has been trans-mitted, the TWINT Flag is set by hardware, and the status code in TWSR will be $08 (See Table 20-2). In order toenter MT mode, SLA+W must be transmitted. This is done by writing SLA+W to TWDR. Thereafter the TWINT bitshould be cleared (by writing it to one) to continue the transfer. This is accomplished by writing the following valueto TWCR:

When SLA+W have been transmitted and an acknowledgement bit has been received, TWINT is set again and anumber of status codes in TWSR are possible. Possible status codes in master mode are $18, $20, or $38. Theappropriate action to be taken for each of these status codes is detailed in Table 20-2.

TWCR TWINT TWEA TWSTA TWSTO TWWC TWEN – TWIE

Value 1 X 1 0 X 1 0 X


Value 1 X 0 0 X 1 0 X

Device 1MASTER

TRANSMITTER

Device 2SLAVE

RECEIVERDevice 3 Device n

SDA

SCL

........ R1 R2

VCC


When SLA+W has been successfully transmitted, a data packet should be transmitted. This is done by writing thedata byte to TWDR. TWDR must only be written when TWINT is high. If not, the access will be discarded, and theWrite Collision bit (TWWC) will be set in the TWCR Register. After updating TWDR, the TWINT bit should becleared (by writing it to one) to continue the transfer. This is accomplished by writing the following value to TWCR:

This scheme is repeated until the last byte has been sent and the transfer is ended by generating a STOP condi-tion or a repeated START condition. A STOP condition is generated by writing the following value to TWCR:

A REPEATED START condition is generated by writing the following value to TWCR:

After a repeated START condition (state $10) the Two-wire Serial Interface can access the same slave again, or anew slave without transmitting a STOP condition. Repeated START enables the master to switch between slaves,master transmitter mode and master receiver mode without losing control of the bus.


Value 1 X 0 0 X 1 0 X


Value 1 X 0 1 X 1 0 X


Value 1 X 1 0 X 1 0 X

Table 21-2. Status Codes for Master Transmitter ModeStatus Code(TWSR)Prescaler Bitsare 0

Status of the Two-wire SerialBus and Two-wire Serial Inter-face Hardware

Application Software Response

Next Action Taken by TWI HardwareTo/from TWDR

To TWCR

STA STO TWINT TWEA

$08 A START condition has beentransmitted

Load SLA+W 0 0 1 X SLA+W will be transmitted;ACK or NOT ACK will be received

$10 A repeated START conditionhas been transmitted

Load SLA+W or

Load SLA+R

0

0

0

0

1

1

X

X

SLA+W will be transmitted;ACK or NOT ACK will be receivedSLA+R will be transmitted;Logic will switch to Master Receiver mode

$18 SLA+W has been transmitted;ACK has been received

Load data byte or

No TWDR action orNo TWDR action or

No TWDR action

0

10

1

0

01

1

1

11

1

X

XX

X

Data byte will be transmitted and ACK or NOT ACK will be receivedRepeated START will be transmittedSTOP condition will be transmitted andTWSTO Flag will be ResetSTOP condition followed by a START condition will be transmitted and TWSTO Flag will be Reset

$20 SLA+W has been transmitted;NOT ACK has been received

Load data byte or


No TWDR action

0

10

1

0

01

1

1

11

1

X

XX

X

Data byte will be transmitted and ACK or NOT ACK will be receivedRepeated START will be transmittedSTOP condition will be transmitted andTWSTO Flag will be resetSTOP condition followed by a START condition will be transmitted and TWSTO Flag will be reset

$28 Data byte has been transmitted;ACK has been received

Load data byte or


No TWDR action

0

10

1

0

01

1

1

11

1

X

XX

X


$30 Data byte has been transmitted;NOT ACK has been received

Load data byte or


No TWDR action

0

10

1

0

01

1

1

11

1

X

XX

X


$38 Arbitration lost in SLA+W or databytes

No TWDR action or

No TWDR action

0

1

0

0

1

1

X

X

Two-wire Serial Bus will be released and not addressed slave mode enteredA START condition will be transmitted when the bus be-comes free


Figure 21-12. Formats and States in the Master Transmitter Mode

21.7.2 Master Receiver ModeIn the Master Receiver mode, a number of data bytes are received from a slave transmitter (see Figure 20-13). Inorder to enter a Master mode, a START condition must be transmitted. The format of the following address packetdetermines whether Master Transmitter or Master Receiver mode is to be entered. If SLA+W is transmitted, MTmode is entered, if SLA+R is transmitted, MR mode is entered. All the status codes mentioned in this sectionassume that the prescaler bits are zero or are masked to zero.

S SLA W A DATA A P

$08 $18 $28

R SLA W

$10

A P

$20

P

$30

A or A

$38

A

Other mastercontinues A or A

$38

Other mastercontinues

R

A

$68


$78 $B0To correspondingstates in slave mode

MT

MR

Successfulltransmissionto a slavereceiver

Next transferstarted with arepeated startcondition

Not acknowledgereceived after theslave address

Not acknowledgereceived after a databyte

Arbitration lost in slaveaddress or data byte

Arbitration lost andaddressed as slave

DATA A

n

From master to slave

From slave to master

Any number of data bytesand their associated acknowledge bits

This number (contained in TWSR) correspondsto a defined state of the Two-wire Serial Bus. Theprescaler bits are zero or masked to zero

S


Figure 21-13. Data Transfer in Master Receiver Mode

A START condition is sent by writing the following value to TWCR:

TWEN must be written to one to enable the Two-wire Serial Interface, TWSTA must be written to one to transmit aSTART condition and TWINT must be set to clear the TWINT Flag. The TWI will then test the Two-wire Serial Busand generate a START condition as soon as the bus becomes free. After a START condition has been transmitted,the TWINT Flag is set by hardware, and the status code in TWSR will be $08 (See Table 20-2). In order to enterMR mode, SLA+R must be transmitted. This is done by writing SLA+R to TWDR. Thereafter the TWINT bit shouldbe cleared (by writing it to one) to continue the transfer. This is accomplished by writing the following value toTWCR:

When SLA+R have been transmitted and an acknowledgement bit has been received, TWINT is set again and anumber of status codes in TWSR are possible. Possible status codes in master mode are $38, $40, or $48. Theappropriate action to be taken for each of these status codes is detailed in Table 20-3. Received data can be readfrom the TWDR Register when the TWINT Flag is set high by hardware. This scheme is repeated until the last bytehas been received. After the last byte has been received, the MR should inform the ST by sending a NACK afterthe last received data byte. The transfer is ended by generating a STOP condition or a repeated START condition.A STOP condition is generated by writing the following value to TWCR:

A REPEATED START condition is generated by writing the following value to TWCR:

After a repeated START condition (state $10) the Two-wire Serial Interface can access the same slave again, or anew slave without transmitting a STOP condition. Repeated START enables the master to switch between slaves,Master Transmitter mode and Master Receiver mode without losing control over the bus.


Value 1 X 1 0 X 1 0 X


Value 1 X 0 0 X 1 0 X


Value 1 X 0 1 X 1 0 X


Value 1 X 1 0 X 1 0 X

Device 1MASTER

RECEIVER

Device 2SLAVE

TRANSMITTERDevice 3 Device n

SDA

SCL

........ R1 R2

VCC

Table 21-3. Status Codes for Master Receiver Mode Status Code(TWSR) Prescaler Bitsare 0




To TWCR

STA STO TWINT TWEA

$08 A START condition has beentransmitted

Load SLA+R 0 0 1 X SLA+R will be transmittedACK or NOT ACK will be received


Figure 21-14. Formats and States in the Master Receiver Mode

21.7.3 Slave Receiver ModeIn the Slave Receiver mode, a number of data bytes are received from a master transmitter (see Figure 20-15). Allthe status codes mentioned in this section assume that the prescaler bits are zero or are masked to zero.

$10 A repeated START conditionhas been transmitted

Load SLA+R or

Load SLA+W

0

0

0

0

1

1

X

X

SLA+R will be transmittedACK or NOT ACK will be receivedSLA+W will be transmittedLogic will switch to masTer Transmitter mode

$38 Arbitration lost in SLA+R or NOTACK bit

No TWDR action or

No TWDR action

0

1

0

0

1

1

X

X

Two-wire Serial Bus will be released and not addressed slave mode will be enteredA START condition will be transmitted when the busbecomes free

$40 SLA+R has been transmitted;ACK has been received

No TWDR action or

No TWDR action

0

0

0

0

1

1

0

1

Data byte will be received and NOT ACK will be returnedData byte will be received and ACK will be returned

$48 SLA+R has been transmitted;NOT ACK has been received


No TWDR action

10

1

01

1

11

1

XX

X

Repeated START will be transmittedSTOP condition will be transmitted and TWSTO Flag will be resetSTOP condition followed by a START condition will be transmitted and TWSTO Flag will be reset

$50 Data byte has been received;ACK has been returned

Read data byte or

Read data byte

0

0

0

0

1

1

0

1


$58 Data byte has been received;NOT ACK has been returned

Read data byte orRead data byte or

Read data byte

10

1

01

1

11

1

XX

X

Repeated START will be transmittedSTOP condition will be transmitted and TWSTO Flag will be resetSTOP condition followed by a START condition will be transmitted and TWSTO Flag will be reset

Table 21-3. Status Codes for Master Receiver Mode (Continued)

S SLA R A DATA A

$08 $40 $50

SLA R

$10

A P

$48

A or A

$38


$38


W

A

$68


$78 $B0To correspondingstates in slave mode

MR

MT

Successfullreceptionfrom a slavereceiver

Next transferstarted with arepeated startcondition

Not acknowledgereceived after theslave address

Arbitration lost in slaveaddress or data byte

Arbitration lost andaddressed as slave

DATA A

n




This number (contained in TWSR) correspondsto a defined state of the Two-wire Serial Bus. The prescaler bits are zero or masked to zero

PDATA A

$58

A

RS


Figure 21-15. Data Transfer in Slave Receiver Mode

To initiate the Slave Receiver mode, TWAR and TWCR must be initialized as follows:

The upper seven bits are the address to which the Two-wire Serial Interface will respond when addressed by amaster. If the LSB is set, the TWI will respond to the general call address ($00), otherwise it will ignore the generalcall address.

TWEN must be written to one to enable the TWI. The TWEA bit must be written to one to enable the acknowledge-ment of the device’s own slave address or the general call address. TWSTA and TWSTO must be written to zero.

When TWAR and TWCR have been initialized, the TWI waits until it is addressed by its own slave address (or thegeneral call address if enabled) followed by the data direction bit. If the direction bit is “0” (write), the TWI will oper-ate in SR mode, otherwise ST mode is entered. After its own slave address and the write bit have been received,the TWINT Flag is set and a valid status code can be read from TWSR. The status code is used to determine theappropriate software action. The appropriate action to be taken for each status code is detailed in Table 20-4. TheSlave Receiver mode may also be entered if arbitration is lost while the TWI is in the Master mode (see states $68and $78).

If the TWEA bit is reset during a transfer, the TWI will return a “Not Acknowledge” (“1”) to SDA after the nextreceived data byte. This can be used to indicate that the slave is not able to receive any more bytes. While TWEAis zero, the TWI does not acknowledge its own slave address. However, the Two-wire Serial Bus is still monitoredand address recognition may resume at any time by setting TWEA. This implies that the TWEA bit may be used totemporarily isolate the TWI from the Two-wire Serial Bus.

In all sleep modes other than Idle Mode, the clock system to the TWI is turned off. If the TWEA bit is set, the inter-face can still acknowledge its own slave address or the general call address by using the Two-wire Serial Bus clockas a clock source. The part will then wake up from sleep and the TWI will hold the SCL clock low during the wakeup and until the TWINT Flag is cleared (by writing it to one). Further data reception will be carried out as normal,with the AVR clocks running as normal. Observe that if the AVR is set up with a long start-up time, the SCL linemay be held low for a long time, blocking other data transmissions.

Note that the Two-wire Serial Interface Data Register – TWDR does not reflect the last byte present on the buswhen waking up from these sleep modes.

TWAR TWA6 TWA5 TWA4 TWA3 TWA2 TWA1 TWA0 TWGCE

Value Device’s Own Slave Address


Value 0 1 0 0 0 1 0 X

Device 3 Device n

SDA

SCL

........ R1 R2

VCC

Device 2MASTER

TRANSMITTER

Device 1SLAVE

RECEIVER


Table 21-4. Status Codes for Slave Receiver Mode Status Code(TWSR)Prescaler Bitsare 0

Status of the Two-wire Serial Busand Two-wire Serial InterfaceHardware



To TWCR

STA STO TWINT TWEA

$60 Own SLA+W has been received;ACK has been returned

No TWDR action or

No TWDR action

X

X

0

0

1

1

0

1


$68 Arbitration lost in SLA+R/W asmaster; own SLA+W has been received; ACK has been returned

No TWDR action or

No TWDR action

X

X

0

0

1

1

0

1


$70 General call address has been received; ACK has been returned

No TWDR action or

No TWDR action

X

X

0

0

1

1

0

1


$78 Arbitration lost in SLA+R/W asmaster; General call address hasbeen received; ACK has been returned

No TWDR action or

No TWDR action

X

X

0

0

1

1

0

1


$80 Previously addressed with ownSLA+W; data has been received;ACK has been returned

Read data byte or

Read data byte

X

X

0

0

1

1

0

1


$88 Previously addressed with ownSLA+W; data has been received;NOT ACK has been returned

Read data byte or

Read data byte or

Read data byte or

Read data byte

0

0

1

1

0

0

0

0

1

1

1

1

0

1

0

1

Switched to the not addressed Slave mode;no recognition of own SLA or GCASwitched to the not addressed Slave mode;own SLA will be recognized;GCA will be recognized if TWGCE = “1”Switched to the not addressed Slave mode;no recognition of own SLA or GCA;a START condition will be transmitted when the bus becomes freeSwitched to the not addressed Slave mode;own SLA will be recognized;GCA will be recognized if TWGCE = “1”;a START condition will be transmitted when the bus becomes free

$90 Previously addressed with general call; data has been re-ceived; ACK has been returned

Read data byte or

Read data byte

X

X

0

0

1

1

0

1


$98 Previously addressed with general call; data has been received; NOT ACK has been returned

Read data byte or

Read data byte or

Read data byte or

Read data byte

0

0

1

1

0

0

0

0

1

1

1

1

0

1

0

1


$A0 A STOP condition or repeatedSTART condition has been received while still addressed asslave

No action 0

0

1

1

0

0

0

0

1

1

1

1

0

1

0

1



Figure 21-16. Formats and States in the Slave Receiver Mode

21.7.4 Slave Transmitter ModeIn the Slave Transmitter mode, a number of data bytes are transmitted to a master receiver (see Figure 20-17). Allthe status codes mentioned in this section assume that the prescaler bits are zero or are masked to zero.

Figure 21-17. Data Transfer in Slave Transmitter Mode

To initiate the Slave Transmitter mode, TWAR and TWCR must be initialized as follows:

S SLA W A DATA A

$60 $80

$88

A

$68

Reception of the ownslave address and one ormore data bytes. All areacknowledged

Last data byte receivedis not acknowledged

Arbitration lost as masterand addressed as slave

Reception of the general calladdress and one or more databytes

Last data byte received isnot acknowledged

n





P or SDATA A

$80 $A0

P or SA

A DATA A

$70 $90

$98

A

$78

P or SDATA A

$90 $A0

P or SA

General Call

Arbitration lost as master andaddressed as slave by general call

DATA A

TWAR TWA6 TWA5 TWA4 TWA3 TWA2 TWA1 TWA0 TWGCE

Value Device’s Own Slave Address

Device 3 Device n

SDA

SCL

........ R1 R2

VCC

Device 2MASTER

RECEIVER

Device 1SLAVE

TRANSMITTER


The upper seven bits are the address to which the Two-wire Serial Interface will respond when addressed by amaster. If the LSB is set, the TWI will respond to the general call address ($00), otherwise it will ignore the generalcall address.

TWEN must be written to one to enable the TWI. The TWEA bit must be written to one to enable the acknowledge-ment of the device’s own slave address or the general call address. TWSTA and TWSTO must be written to zero.

When TWAR and TWCR have been initialized, the TWI waits until it is addressed by its own slave address (or thegeneral call address if enabled) followed by the data direction bit. If the direction bit is “1” (read), the TWI will oper-ate in ST mode, otherwise SR mode is entered. After its own slave address and the write bit have been received,the TWINT Flag is set and a valid status code can be read from TWSR. The status code is used to determine theappropriate software action. The appropriate action to be taken for each status code is detailed in Table 20-5. Theslave transmitter mode may also be entered if arbitration is lost while the TWI is in the Master mode (see state$B0).

If the TWEA bit is written to zero during a transfer, the TWI will transmit the last byte of the transfer. State $C0 orstate $C8 will be entered, depending on whether the master receiver transmits a NACK or ACK after the final byte.The TWI is switched to the not addressed Slave mode, and will ignore the master if it continues the transfer. Thusthe master receiver receives all “1” as serial data. State $C8 is entered if the master demands additional data bytes(by transmitting ACK), even though the slave has transmitted the last byte (TWEA zero and expecting NACK fromthe master).

While TWEA is zero, the TWI does not respond to its own slave address. However, the Two-wire Serial Bus is stillmonitored and address recognition may resume at any time by setting TWEA. This implies that the TWEA bit maybe used to temporarily isolate the TWI from the Two-wire Serial Bus.

In all sleep modes other than Idle mode, the clock system to the TWI is turned off. If the TWEA bit is set, the inter-face can still acknowledge its own slave address or the general call address by using the Two-wire Serial Bus clockas a clock source. The part will then wake up from sleep and the TWI will hold the SCL clock will low during thewake up and until the TWINT Flag is cleared (by writing it to one). Further data transmission will be carried out asnormal, with the AVR clocks running as normal. Observe that if the AVR is set up with a long start-up time, the SCLline may be held low for a long time, blocking other data transmissions.

Note that the Two-wire Serial Interface Data Register – TWDR does not reflect the last byte present on the buswhen waking up from these sleep modes.


Value 0 1 0 0 0 1 0 X


Figure 21-18. Formats and States in the Slave Transmitter Mode

Table 21-5. Status Codes for Slave Transmitter ModeStatus Code(TWSR)Prescaler Bitsare 0

Status of the Two-wire Serial Busand Two-wire Serial InterfaceHardware



To TWCR

STA STO TWINT TWEA

$A8 Own SLA+R has been received;ACK has been returned

Load data byte or

Load data byte

X

X

0

0

1

1

0

1

Last data byte will be transmitted and NOT ACK should be receivedData byte will be transmitted and ACK should be re-ceived

$B0 Arbitration lost in SLA+R/W asmaster; own SLA+R has been received; ACK has been returned

Load data byte or

Load data byte

X

X

0

0

1

1

0

1


$B8 Data byte in TWDR has been transmitted; ACK has been received

Load data byte or

Load data byte

X

X

0

0

1

1

0

1


$C0 Data byte in TWDR has been transmitted; NOT ACK has been received

No TWDR action or

No TWDR action or

No TWDR action or

No TWDR action

0

0

1

1

0

0

0

0

1

1

1

1

0

1

0

1


$C8 Last data byte in TWDR has beentransmitted (TWEA = “0”); ACKhas been received

No TWDR action or

No TWDR action or

No TWDR action or

No TWDR action

0

0

1

1

0

0

0

0

1

1

1

1

0

1

0

1


S SLA R A DATA A

$A8 $B8

A

$B0

Reception of the ownslave address and one ormore data bytes

Last data byte transmitted.Switched to not addressedslave (TWEA = '0')

Arbitration lost as masterand addressed as slave

n





P or SDATA

$C0

DATA A

A

$C8

P or SAll 1's

A


21.7.5 Miscellaneous StatesThere are two status codes that do not correspond to a defined TWI state, see Table 20-6.

Status $F8 indicates that no relevant information is available because the TWINT Flag is not set. This occursbetween other states, and when the TWI is not involved in a serial transfer.

Status $00 indicates that a bus error has occurred during a Two-wire Serial Bus transfer. A bus error occurs whena START or STOP condition occurs at an illegal position in the format frame. Examples of such illegal positions areduring the serial transfer of an address byte, a data byte, or an acknowledge bit. When a bus error occurs, TWINTis set. To recover from a bus error, the TWSTO Flag must set and TWINT must be cleared by writing a logic one toit. This causes the TWI to enter the not addressed slave mode and to clear the TWSTO Flag (no other bits inTWCR are affected). The SDA and SCL lines are released, and no STOP condition is transmitted.

21.7.6 Combining Several TWI ModesIn some cases, several TWI modes must be combined in order to complete the desired action. Consider for exam-ple reading data from a serial EEPROM. Typically, such a transfer involves the following steps:

1. The transfer must be initiated

2. The EEPROM must be instructed what location should be read

3. The reading must be performed

4. The transfer must be finished

Note that data is transmitted both from master to slave and vice versa. The master must instruct the slave whatlocation it wants to read, requiring the use of the MT mode. Subsequently, data must be read from the slave, imply-ing the use of the MR mode. Thus, the transfer direction must be changed. The master must keep control of thebus during all these steps, and the steps should be carried out as an atomical operation. If this principle is violatedin a multimaster system, another master can alter the data pointer in the EEPROM between steps 2 and 3, and themaster will read the wrong data location. Such a change in transfer direction is accomplished by transmitting aREPEATED START between the transmission of the address byte and reception of the data. After a REPEATEDSTART, the master keeps ownership of the bus. The following figure shows the flow in this transfer.

Figure 21-19. Combining Several TWI Modes to Access a Serial EEPROM

21.8 Multi-master Systems and ArbitrationIf multiple masters are connected to the same bus, transmissions may be initiated simultaneously by one or moreof them. The TWI standard ensures that such situations are handled in such a way that one of the masters will be

Table 21-6. Miscellaneous StatesStatus Code(TWSR)Prescaler Bitsare 0




To TWCR

STA STO TWINT TWEA

$F8 No relevant state informationavailable; TWINT = “0”

No TWDR action No TWCR action Wait or proceed current transfer

$00 Bus error due to an illegalSTART or STOP condition

No TWDR action 0 1 1 X Only the internal hardware is affected, no STOP condi-tion is sent on the bus. In all cases, the bus is released and TWSTO is cleared.

Master Transmitter Master Receiver

S = START Rs = REPEATED START P = STOP

Transmitted from Master to Slave Transmitted from Slave to Master

S SLA+W A ADDRESS A Rs SLA+R A DATA A P


allowed to proceed with the transfer, and that no data will be lost in the process. An example of an arbitration situ-ation is depicted below, where two masters are trying to transmit data to a slave receiver.

Figure 21-20. An Arbitration Example

Device 1MASTER

TRANSMITTER

Device 2MASTER

TRANSMITTER

Device 3SLAVE

RECEIVERDevice n

SDA

SCL

........ R1 R2

VCC


Several different scenarios may arise during arbitration, as described below:

• Two or more masters are performing identical communication with the same slave. In this case, neither the slave nor any of the masters will know about the bus contention.

• Two or more masters are accessing the same slave with different data or direction bit. In this case, arbitration will occur, either in the READ/WRITE bit or in the data bits. The masters trying to output a one on SDA while another master outputs a zero will lose the arbitration. Losing masters will switch to not addressed slave mode or wait until the bus is free and transmit a new START condition, depending on application software action.

• Two or more masters are accessing different slaves. In this case, arbitration will occur in the SLA bits. Masters trying to output a one on SDA while another master outputs a zero will lose the arbitration. Masters losing arbitration in SLA will switch to slave mode to check if they are being addressed by the winning master. If addressed, they will switch to SR or ST mode, depending on the value of the READ/WRITE bit. If they are not being addressed, they will switch to not addressed slave mode or wait until the bus is free and transmit a new START condition, depending on application software action.

This is summarized in Figure 20-21. Possible status values are given in circles.

Figure 21-21. Possible Status Codes Caused by Arbitration

OwnAddress / General Call

received

Arbitration lost in SLA

TWI bus will be released and not addressed slave mode will be enteredA START condition will be transmitted when the bus becomes free

No

Arbitration lost in Data

Direction

Yes

Write Data byte will be received and NOT ACK will be returnedData byte will be received and ACK will be returned

Last data byte will be transmitted and NOT ACK should be receivedData byte will be transmitted and ACK should be received

ReadB0

68/78

38

SLASTART Data STOP



21.9.1 TWBR – TWI Bit Rate Register

• Bits 7:0 – TWI Bit Rate RegisterTWBR selects the division factor for the bit rate generator. The bit rate generator is a frequency divider which gen-erates the SCL clock frequency in the Master modes. See “Bit Rate Generator Unit” on page 182 for calculating bitrates.

21.9.2 TWCR – TWI Control RegisterThe TWCR is used to control the operation of the TWI. It is used to enable the TWI, to initiate a master access byapplying a START condition to the bus, to generate a receiver acknowledge, to generate a stop condition, and tocontrol halting of the bus while the data to be written to the bus are written to the TWDR. It also indicates a writecollision if data is attempted written to TWDR while the register is inaccessible.

• Bit 7 – TWINT: TWI Interrupt FlagThis bit is set by hardware when the TWI has finished its current job and expects application software response. Ifthe I-bit in SREG and TWIE in TWCR are set, the MCU will jump to the TWI Interrupt Vector. While the TWINT Flagis set, the SCL low period is stretched.

The TWINT Flag must be cleared by software by writing a logic one to it. Note that this flag is not automaticallycleared by hardware when executing the interrupt routine. Also note that clearing this flag starts the operation ofthe TWI, so all accesses to the TWI Address Register (TWAR), TWI Status Register (TWSR), and TWI Data Reg-ister (TWDR) must be complete before clearing this flag.

• Bit 6 – TWEA: TWI Enable Acknowledge BitThe TWEA bit controls the generation of the acknowledge pulse. If the TWEA bit is written to one, the ACK pulse isgenerated on the TWI bus if the following conditions are met:

1. The device’s own slave address has been received.

2. A general call has been received, while the TWGCE bit in the TWAR is set.

3. A data byte has been received in Master Receiver or Slave Receiver mode.

By writing the TWEA bit to zero, the device can be virtually disconnected from the Two-wire Serial Bus temporarily.Address recognition can then be resumed by writing the TWEA bit to one again.

• Bit 5 – TWSTA: TWI START Condition BitThe application writes the TWSTA bit to one when it desires to become a master on the Two-wire Serial Bus. TheTWI hardware checks if the bus is available, and generates a START condition on the bus if it is free. However, ifthe bus is not free, the TWI waits until a STOP condition is detected, and then generates a new START condition toclaim the bus Master status. TWSTA must be cleared by software when the START condition has beentransmitted.

Bit 7 6 5 4 3 2 1 0

TWBR7 TWBR6 TWBR5 TWBR4 TWBR3 TWBR2 TWBR1 TWBR0 TWBR



Bit 7 6 5 4 3 2 1 0

TWINT TWEA TWSTA TWSTO TWWC TWEN – TWIE TWCR

Read/Write R/W R/W R/W R/W R R/W R R/W



• Bit 4 – TWSTO: TWI STOP Condition BitWriting the TWSTO bit to one in Master mode will generate a STOP condition on the Two-wire Serial Bus. Whenthe STOP condition is executed on the bus, the TWSTO bit is cleared automatically. In slave mode, setting theTWSTO bit can be used to recover from an error condition. This will not generate a STOP condition, but the TWIreturns to a well-defined unaddressed slave mode and releases the SCL and SDA lines to a high impedance state.

• Bit 3 – TWWC: TWI Write Collision FlagThe TWWC bit is set when attempting to write to the TWI Data Register – TWDR when TWINT is low. This flag iscleared by writing the TWDR Register when TWINT is high.

• Bit 2 – TWEN: TWI Enable BitThe TWEN bit enables TWI operation and activates the TWI interface. When TWEN is written to one, the TWItakes control over the I/O pins connected to the SCL and SDA pins, enabling the slew-rate limiters and spike filters.If this bit is written to zero, the TWI is switched off and all TWI transmissions are terminated, regardless of anyongoing operation.

• Bit 1 – ReservedThis bit is a reserved bit and will always read as zero.

• Bit 0 – TWIE: TWI Interrupt EnableWhen this bit is written to one, and the I-bit in SREG is set, the TWI interrupt request will be activated for as long asthe TWINT Flag is high.

21.9.3 TWSR – TWI Status Register

• Bits 7:3 – TWS: TWI StatusThese five bits reflect the status of the TWI logic and the Two-wire Serial Bus. The different status codes aredescribed later in this section. Note that the value read from TWSR contains both the 5-bit status value and the 2-bit prescaler value. The application designer should mask the prescaler bits to zero when checking the Status bits.This makes status checking independent of prescaler setting. This approach is used in this datasheet, unless oth-erwise noted.

• Bit 2 – ReservedThis bit is reserved and will always read as zero.

• Bits 1:0 – TWPS: TWI Prescaler BitsThese bits can be read and written, and control the bit rate prescaler.

Bit 7 6 5 4 3 2 1 0

TWS7 TWS6 TWS5 TWS4 TWS3 – TWPS1 TWPS0 TWSR

Read/Write R R R R R R R/W R/W


Table 21-7. TWI Bit Rate Prescaler

TWPS1 TWPS0 Prescaler Value

0 0 1

0 1 4

1 0 16

1 1 64


To calculate bit rates, see “Bit Rate Generator Unit” on page 182. The value of TWPS1:0 is used in the equation.

21.9.4 TWDR – TWI Data Register

In Transmit mode, TWDR contains the next byte to be transmitted. In Receive mode, the TWDR contains the lastbyte received. It is writable while the TWI is not in the process of shifting a byte. This occurs when the TWI InterruptFlag (TWINT) is set by hardware. Note that the Data Register cannot be initialized by the user before the first inter-rupt occurs. The data in TWDR remains stable as long as TWINT is set. While data is shifted out, data on the busis simultaneously shifted in. TWDR always contains the last byte present on the bus, except after a wake up from asleep mode by the TWI interrupt. In this case, the contents of TWDR is undefined. In the case of a lost bus arbitra-tion, no data is lost in the transition from Master to Slave. Handling of the ACK bit is controlled automatically by theTWI logic, the CPU cannot access the ACK bit directly.

• Bits 7:0 – TWD: TWI Data Register These eight bits contin the next data byte to be transmitted, or the latest data byte received on the Two-wire SerialBus.

21.9.5 TWAR – TWI (Slave) Address Register

The TWAR should be loaded with the 7-bit slave address (in the seven most significant bits of TWAR) to which theTWI will respond when programmed as a slave transmitter or receiver. In multimaster systems, TWAR must be setin masters which can be addressed as slaves by other masters.

The LSB of TWAR is used to enable recognition of the general call address ($00). There is an associated addresscomparator that looks for the slave address (or general call address if enabled) in the received serial address. If amatch is found, an interrupt request is generated.

• Bits 7:1 – TWA: TWI (Slave) Address Register These seven bits constitute the slave address of the TWI unit.

• Bit 0 – TWGCE: TWI General Call Recognition Enable Bit If set, this bit enables the recognition of a General Call given over the Two-wire Serial Bus.

Bit 7 6 5 4 3 2 1 0

TWD7 TWD6 TWD5 TWD4 TWD3 TWD2 TWD1 TWD0 TWDR



Bit 7 6 5 4 3 2 1 0

TWA6 TWA5 TWA4 TWA3 TWA2 TWA1 TWA0 TWGCE TWAR




22. Analog Comparator

22.1 OverviewThe Analog Comparator compares the input values on the positive pin AIN0 and negative pin AIN1. When the volt-age on the positive pin AIN0 is higher than the voltage on the negative pin AIN1, the Analog Comparator Output,ACO, is set. The comparator’s output can be set to trigger the Timer/Counter1 Input Capture function. In addition,the comparator can trigger a separate interrupt, exclusive to the Analog Comparator. The user can select Interrupttriggering on comparator output rise, fall or toggle. A block diagram of the comparator and its surrounding logic isshown in Figure 21-1.

Figure 22-1. Analog Comparator Block Diagram(2)

Notes: 1. See Table 21-1 on page 206.2. Refer to Figure 1-1 on page 2 and Table 12-6 on page 59 for Analog Comparator pin placement.

22.2 Analog Comparator Multiplexed InputIt is possible to select any of the ADC7:0 pins to replace the negative input to the Analog Comparator. The ADCmultiplexer is used to select this input, and consequently, the ADC must be switched off to utilize this feature. If theAnalog Comparator Multiplexer Enable bit (ACME in SFIOR) is set and the ADC is switched off (ADEN in ADCSRAis zero), MUX2:0 in ADMUX select the input pin to replace the negative input to the Analog Comparator, as shownin Table 21-1. If ACME is cleared or ADEN is set, AIN1 is applied to the negative input to the Analog Comparator.

ACBG

BANDGAPREFERENCE

ADC MULTIPLEXEROUTPUT

ACMEADEN

(1)

Table 22-1. Analog Comparator Multiplexed Input

ACME ADEN MUX2:0 Analog Comparator Negative Input

0 x xxx AIN1

1 1 xxx AIN1

1 0 000 ADC0

1 0 001 ADC1

1 0 010 ADC2




• Bit 3 – ACME: Analog Comparator Multiplexer EnableWhen this bit is written logic one and the ADC is switched off (ADEN in ADCSRA is zero), the ADC multiplexerselects the negative input to the Analog Comparator. When this bit is written logic zero, AIN1 is applied to the neg-ative input of the Analog Comparator. For a detailed description of this bit, see “Analog Comparator MultiplexedInput” on page 205.

22.3.2 ACSR – Analog Comparator Control and Status Register

• Bit 7 – ACD: Analog Comparator DisableWhen this bit is written logic one, the power to the Analog Comparator is switched off. This bit can be set at anytime to turn off the Analog Comparator. This will reduce power consumption in active and Idle mode. When chang-ing the ACD bit, the Analog Comparator Interrupt must be disabled by clearing the ACIE bit in ACSR. Otherwise aninterrupt can occur when the bit is changed.

• Bit 6 – ACBG: Analog Comparator Bandgap SelectWhen this bit is set, a fixed bandgap reference voltage replaces the positive input to the Analog Comparator. Whenthis bit is cleared, AIN0 is applied to the positive input of the Analog Comparator. See “Internal Voltage Reference”on page 41.

• Bit 5 – ACO: Analog Comparator OutputThe output of the Analog Comparator is synchronized and then directly connected to ACO. The synchronizationintroduces a delay of 1 - 2 clock cycles.

• Bit 4 – ACI: Analog Comparator Interrupt FlagThis bit is set by hardware when a comparator output event triggers the interrupt mode defined by ACIS1 andACIS0. The Analog Comparator Interrupt routine is executed if the ACIE bit is set and the I-bit in SREG is set. ACIis cleared by hardware when executing the corresponding interrupt handling vector. Alternatively, ACI is cleared bywriting a logic one to the flag.

1 0 011 ADC3

1 0 100 ADC4

1 0 101 ADC5

1 0 110 ADC6

1 0 111 ADC7

Table 22-1. Analog Comparator Multiplexed Input (Continued)

ACME ADEN MUX2:0 Analog Comparator Negative Input

Bit 7 6 5 4 3 2 1 0




Bit 7 6 5 4 3 2 1 0

ACD ACBG ACO ACI ACIE ACIC ACIS1 ACIS0 ACSR


Initial Value 0 0 N/A 0 0 0 0 0


• Bit 3 – ACIE: Analog Comparator Interrupt EnableWhen the ACIE bit is written logic one and the I-bit in the Status Register is set, the Analog Comparator Interrupt isactivated. When written logic zero, the interrupt is disabled.

• Bit 2 – ACIC: Analog Comparator Input Capture EnableWhen written logic one, this bit enables the Input Capture function in Timer/Counter1 to be triggered by the AnalogComparator. The comparator output is in this case directly connected to the Input Capture front-end logic, makingthe comparator utilize the noise canceler and edge select features of the Timer/Counter1 Input Capture interrupt.When written logic zero, no connection between the Analog Comparator and the Input Capture function exists. Tomake the comparator trigger the Timer/Counter1 Input Capture interrupt, the TICIE1 bit in the Timer Interrupt MaskRegister (TIMSK) must be set.

• Bits 1:0 – ACIS[1:0]: Analog Comparator Interrupt Mode SelectThese bits determine which comparator events that trigger the Analog Comparator interrupt. The different settingsare shown in Table 21-2.

When changing the ACIS1/ACIS0 bits, the Analog Comparator Interrupt must be disabled by clearing its InterruptEnable bit in the ACSR Register. Otherwise an interrupt can occur when the bits are changed.

Table 22-2. ACIS1/ACIS0 Settings

ACIS1 ACIS0 Interrupt Mode

0 0 Comparator Interrupt on Output Toggle

0 1 Reserved

1 0 Comparator Interrupt on Falling Output Edge

1 1 Comparator Interrupt on Rising Output Edge


23. Analog to Digital Converter

23.1 Features• 10-bit Resolution• 0.5 LSB Integral Non-linearity• ±2 LSB Absolute Accuracy• 13 - 260 µs Conversion Time• Up to 15kSPS at Maximum Resolution• 8 Multiplexed Single Ended Input Channels• 7 Differential Input Channels• 2 Differential Input Channels with Optional Gain of 10x and 200x• Optional Left adjustment for ADC Result Readout• 0 - VCC ADC Input Voltage Range• Selectable 2.56V ADC Reference Voltage• Free Running or Single Conversion Mode• ADC Start Conversion by Auto Triggering on Interrupt Sources• Interrupt on ADC Conversion Complete• Sleep Mode Noise Canceler

23.2 OverviewThe ATmega32A features a 10-bit successive approximation ADC. The ADC is connected to an 8-channel AnalogMultiplexer which allows 8 single-ended voltage inputs constructed from the pins of Port A. The single-ended volt-age inputs refer to 0V (GND).

The device also supports 16 differential voltage input combinations. Two of the differential inputs (ADC1, ADC0and ADC3, ADC2) are equipped with a programmable gain stage, providing amplification steps of 0 dB (1x), 20 dB(10x), or 46 dB (200x) on the differential input voltage before the A/D conversion. Seven differential analog inputchannels share a common negative terminal (ADC1), while any other ADC input can be selected as the positiveinput terminal. If 1x or 10x gain is used, 8-bit resolution can be expected. If 200x gain is used, 7-bit resolution canbe expected.

The ADC contains a Sample and Hold circuit which ensures that the input voltage to the ADC is held at a constantlevel during conversion. A block diagram of the ADC is shown in Figure 22-1.

The ADC has a separate analog supply voltage pin, AVCC. AVCC must not differ more than ±0.3V from VCC. Seethe paragraph “ADC Noise Canceler” on page 216 on how to connect this pin.

Internal reference voltages of nominally 2.56V or AVCC are provided On-chip. The voltage reference may be exter-nally decoupled at the AREF pin by a capacitor for better noise performance.


Figure 23-1. Analog to Digital Converter Block Schematic

23.3 OperationThe ADC converts an analog input voltage to a 10-bit digital value through successive approximation. The mini-mum value represents GND and the maximum value represents the voltage on the AREF pin minus 1 LSB.Optionally, AVCC or an internal 2.56V reference voltage may be connected to the AREF pin by writing to theREFSn bits in the ADMUX Register. The internal voltage reference may thus be decoupled by an external capaci-tor at the AREF pin to improve noise immunity.

The analog input channel and differential gain are selected by writing to the MUX bits in ADMUX. Any of the ADCinput pins, as well as GND and a fixed bandgap voltage reference, can be selected as single ended inputs to theADC. A selection of ADC input pins can be selected as positive and negative inputs to the differential gainamplifier.

If differential channels are selected, the differential gain stage amplifies the voltage difference between theselected input channel pair by the selected gain factor. This amplified value then becomes the analog input to theADC. If single ended channels are used, the gain amplifier is bypassed altogether.

ADC CONVERSIONCOMPLETE IRQ

8-BIT DATA BUS

15 0

ADC MULTIPLEXERSELECT (ADMUX)

ADC CTRL. & STATUSREGISTER (ADCSRA)

ADC DATA REGISTER(ADCH/ADCL)

MU

X2

AD

IE

AD

AT

E

AD

SC

AD

EN

AD

IFA

DIF

MU

X1

MU

X0

AD

PS

0

AD

PS

1

AD

PS

2

MU

X3

CONVERSION LOGIC

10-BIT DAC

+-

SAMPLE & HOLDCOMPARATOR

INTERNAL 2.56V REFERENCE

MUX DECODER

MU

X4

AVCC

ADC7

ADC6

ADC5

ADC4

ADC3

ADC2

ADC1

ADC0

RE

FS

0

RE

FS

1

AD

LAR

+

-

CH

AN

NE

L S

ELE

CT

ION

GA

IN S

ELE

CT

ION

AD

C[9

:0]


GAINAMPLIFIER

AREF

BANDGAPREFERENCE

PRESCALER

SINGLE ENDED / DIFFERENTIAL SELECTION

GND

POS.INPUTMUX

NEG.INPUTMUX

TRIGGERSELECT

ADTS[2:0]

INTERRUPTFLAGS

START


The ADC is enabled by setting the ADC Enable bit, ADEN in ADCSRA. Voltage reference and input channel selec-tions will not go into effect until ADEN is set. The ADC does not consume power when ADEN is cleared, so it isrecommended to switch off the ADC before entering power saving sleep modes.

The ADC generates a 10-bit result which is presented in the ADC Data Registers, ADCH and ADCL. By default,the result is presented right adjusted, but can optionally be presented left adjusted by setting the ADLAR bit inADMUX.

If the result is left adjusted and no more than 8-bit precision is required, it is sufficient to read ADCH. Otherwise,ADCL must be read first, then ADCH, to ensure that the content of the Data Registers belongs to the same conver-sion. Once ADCL is read, ADC access to Data Registers is blocked. This means that if ADCL has been read, anda conversion completes before ADCH is read, neither register is updated and the result from the conversion is lost.When ADCH is read, ADC access to the ADCH and ADCL Registers is re-enabled.

The ADC has its own interrupt which can be triggered when a conversion completes. When ADC access to theData Registers is prohibited between reading of ADCH and ADCL, the interrupt will trigger even if the result is lost.

23.4 Starting a ConversionA single conversion is started by writing a logical one to the ADC Start Conversion bit, ADSC. This bit stays high aslong as the conversion is in progress and will be cleared by hardware when the conversion is completed. If a differ-ent data channel is selected while a conversion is in progress, the ADC will finish the current conversion beforeperforming the channel change.

Alternatively, a conversion can be triggered automatically by various sources. Auto Triggering is enabled by settingthe ADC Auto Trigger Enable bit, ADATE in ADCSRA. The trigger source is selected by setting the ADC TriggerSelect bits, ADTS in SFIOR (see description of the ADTS bits for a list of the trigger sources). When a positiveedge occurs on the selected trigger signal, the ADC prescaler is reset and a conversion is started. This provides amethod of starting conversions at fixed intervals. If the trigger signal still is set when the conversion completes, anew conversion will not be started. If another positive edge occurs on the trigger signal during conversion, the edgewill be ignored. Note that an Interrupt Flag will be set even if the specific interrupt is disabled or the global interruptenable bit in SREG is cleared. A conversion can thus be triggered without causing an interrupt. However, the Inter-rupt Flag must be cleared in order to trigger a new conversion at the next interrupt event.

Figure 23-2. ADC Auto Trigger Logic

Using the ADC Interrupt Flag as a trigger source makes the ADC start a new conversion as soon as the ongoingconversion has finished. The ADC then operates in Free Running mode, constantly sampling and updating theADC Data Register. The first conversion must be started by writing a logical one to the ADSC bit in ADCSRA. In

ADSC

ADIF

SOURCE 1

SOURCE n

ADTS[2:0]

CONVERSIONLOGIC

PRESCALER

START CLKADC

.

.

.

. EDGEDETECTOR

ADATE


this mode the ADC will perform successive conversions independently of whether the ADC Interrupt Flag, ADIF iscleared or not.

If Auto Triggering is enabled, single conversions can be started by writing ADSC in ADCSRA to one. ADSC canalso be used to determine if a conversion is in progress. The ADSC bit will be read as one during a conversion,independently of how the conversion was started.

23.5 Prescaling and Conversion Timing

Figure 23-3. ADC Prescaler

By default, the successive approximation circuitry requires an input clock frequency between 50kHz and 200kHz toget maximum resolution. If a lower resolution than 10 bits is needed, the input clock frequency to the ADC can behigher than 200kHz to get a higher sample rate.

The ADC module contains a prescaler, which generates an acceptable ADC clock frequency from any CPU fre-quency above 100kHz. The prescaling is set by the ADPS bits in ADCSRA. The prescaler starts counting from themoment the ADC is switched on by setting the ADEN bit in ADCSRA. The prescaler keeps running for as long asthe ADEN bit is set, and is continuously reset when ADEN is low.

When initiating a single ended conversion by setting the ADSC bit in ADCSRA, the conversion starts at the follow-ing rising edge of the ADC clock cycle. See “Differential Gain Channels” on page 214 for details on differentialconversion timing.

A normal conversion takes 13 ADC clock cycles. The first conversion after the ADC is switched on (ADEN in ADC-SRA is set) takes 25 ADC clock cycles in order to initialize the analog circuitry.

The actual sample-and-hold takes place 1.5 ADC clock cycles after the start of a normal conversion and 13.5 ADCclock cycles after the start of a first conversion. When a conversion is complete, the result is written to the ADCData Registers, and ADIF is set. In single conversion mode, ADSC is cleared simultaneously. The software maythen set ADSC again, and a new conversion will be initiated on the first rising ADC clock edge.

When Auto Triggering is used, the prescaler is reset when the trigger event occurs. This assures a fixed delay fromthe trigger event to the start of conversion. In this mode, the sample-and-hold takes place 2 ADC clock cycles afterthe rising edge on the trigger source signal. Three additional CPU clock cycles are used for synchronization logic.

When using Differential mode, along with Auto Trigging from a source other than the ADC Conversion Complete,each conversion will require 25 ADC clocks. This is because the ADC must be disabled and re-enabled after everyconversion.

7-BIT ADC PRESCALER

ADC CLOCK SOURCE

CK

ADPS0ADPS1ADPS2

CK

/128

CK

/2

CK

/4

CK

/8

CK

/16

CK

/32

CK

/64

ResetADENSTART


In Free Running mode, a new conversion will be started immediately after the conversion completes, while ADSCremains high. For a summary of conversion times, see Table 22-1.

Figure 23-4. ADC Timing Diagram, First Conversion (Single Conversion Mode)

Figure 23-5. ADC Timing Diagram, Single Conversion

Figure 23-6. ADC Timing Diagram, Auto Triggered Conversion

MSB of Result

LSB of Result

ADC Clock

ADSC

Sample & Hold

ADIF

ADCH

ADCL

Cycle Number

ADEN

1 2 12 13 14 15 16 17 18 19 20 21 22 23 24 25 1 2

First ConversionNextConversion

3

MUX and REFSUpdate MUX and REFS

Update

ConversionComplete

1 2 3 4 5 6 7 8 9 10 11 12 13

MSB of Result

LSB of Result

ADC Clock

ADSC

ADIF

ADCH

ADCL

Cycle Number 1 2

One Conversion Next Conversion

3

Sample & HoldMUX and REFSUpdate

ConversionComplete MUX and REFS

Update

1 2 3 4 5 6 7 8 9 10 11 12 13

MSB of Result

LSB of Result

ADC Clock

TriggerSource

ADIF

ADCH

ADCL

Cycle Number 1 2


ConversionComplete

Prescaler Reset

ADATE

PrescalerReset

Sample & Hold

MUX and REFS Update


Figure 23-7. ADC Timing Diagram, Free Running Conversion

23.5.1 Differential Gain ChannelsWhen using differential gain channels, certain aspects of the conversion need to be taken into consideration.

Differential conversions are synchronized to the internal clock CKADC2 equal to half the ADC clock. This synchroni-zation is done automatically by the ADC interface in such a way that the sample-and-hold occurs at a specificphase of CKADC2. A conversion initiated by the user (that is, all single conversions, and the first free running con-version) when CKADC2 is low will take the same amount of time as a single ended conversion (13 ADC clock cyclesfrom the next prescaled clock cycle). A conversion initiated by the user when CKADC2 is high will take 14 ADC clockcycles due to the synchronization mechanism. In Free Running mode, a new conversion is initiated immediatelyafter the previous conversion completes, and since CKADC2 is high at this time, all automatically started (that is, allbut the first) free running conversions will take 14 ADC clock cycles.

The gain stage is optimized for a bandwidth of 4kHz at all gain settings. Higher frequencies may be subjected tonon-linear amplification. An external low-pass filter should be used if the input signal contains higher frequencycomponents than the gain stage bandwidth. Note that the ADC clock frequency is independent of the gain stagebandwidth limitation. For example, the ADC clock period may be 6 µs, allowing a channel to be sampled at12kSPS, regardless of the bandwidth of this channel.

If differential gain channels are used and conversions are started by Auto Triggering, the ADC must be switched offbetween conversions. When Auto Triggering is used, the ADC prescaler is reset before the conversion is started.Since the gain stage is dependent of a stable ADC clock prior to the conversion, this conversion will not be valid.By disabling and then re-enabling the ADC between each conversion (writing ADEN in ADCSRA to “0” then to “1”),only extended conversions are performed. The result from the extended conversions will be valid. See “Prescalingand Conversion Timing” on page 211 for timing details.

Table 23-1. ADC Conversion Time

ConditionSample & Hold (Cycles

from Start of Conversion) Conversion Time (Cycles)

First conversion 13.5 25

Normal conversions, single ended 1.5 13

Auto Triggered conversions 2 13.5

Normal conversions, differential 1.5/2.5 13/14

11 12 13

MSB of Result

LSB of Result

ADC Clock

ADSC

ADIF

ADCH

ADCL

Cycle Number1 2


3 4

ConversionComplete

Sample & Hold

MUX and REFSUpdate


23.6 Changing Channel or Reference SelectionThe MUXn and REFS1:0 bits in the ADMUX Register are single buffered through a temporary register to which theCPU has random access. This ensures that the channels and reference selection only takes place at a safe pointduring the conversion. The channel and reference selection is continuously updated until a conversion is started.Once the conversion starts, the channel and reference selection is locked to ensure a sufficient sampling time forthe ADC. Continuous updating resumes in the last ADC clock cycle before the conversion completes (ADIF inADCSRA is set). Note that the conversion starts on the following rising ADC clock edge after ADSC is written. Theuser is thus advised not to write new channel or reference selection values to ADMUX until one ADC clock cycleafter ADSC is written.

If Auto Triggering is used, the exact time of the triggering event can be indeterministic. Special care must be takenwhen updating the ADMUX Register, in order to control which conversion will be affected by the new settings.

If both ADATE and ADEN is written to one, an interrupt event can occur at any time. If the ADMUX Register ischanged in this period, the user cannot tell if the next conversion is based on the old or the new settings. ADMUXcan be safely updated in the following ways:

1. When ADATE or ADEN is cleared.

2. During conversion, minimum one ADC clock cycle after the trigger event.

3. After a conversion, before the Interrupt Flag used as trigger source is cleared.

When updating ADMUX in one of these conditions, the new settings will affect the next ADC conversion.

Special care should be taken when changing differential channels. Once a differential channel has been selected,the gain stage may take as much as 125 µs to stabilize to the new value. Thus conversions should not be startedwithin the first 125 µs after selecting a new differential channel. Alternatively, conversion results obtained withinthis period should be discarded.

The same settling time should be observed for the first differential conversion after changing ADC reference (bychanging the REFS1:0 bits in ADMUX).

23.6.1 ADC Input ChannelsWhen changing channel selections, the user should observe the following guidelines to ensure that the correctchannel is selected:

In Single Conversion mode, always select the channel before starting the conversion. The channel selection maybe changed one ADC clock cycle after writing one to ADSC. However, the simplest method is to wait for the con-version to complete before changing the channel selection.

In Free Running mode, always select the channel before starting the first conversion. The channel selection maybe changed one ADC clock cycle after writing one to ADSC. However, the simplest method is to wait for the firstconversion to complete, and then change the channel selection. Since the next conversion has already startedautomatically, the next result will reflect the previous channel selection. Subsequent conversions will reflect thenew channel selection.

When switching to a differential gain channel, the first conversion result may have a poor accuracy due to therequired settling time for the automatic offset cancellation circuitry. The user should preferably disregard the firstconversion result.

23.6.2 ADC Voltage ReferenceThe reference voltage for the ADC (VREF) indicates the conversion range for the ADC. Single ended channels thatexceed VREF will result in codes close to 0x3FF. VREF can be selected as either AVCC, internal 2.56V reference, orexternal AREF pin.


AVCC is connected to the ADC through a passive switch. The internal 2.56V reference is generated from the inter-nal bandgap reference (VBG) through an internal amplifier. In either case, the external AREF pin is directlyconnected to the ADC, and the reference voltage can be made more immune to noise by connecting a capacitorbetween the AREF pin and ground. VREF can also be measured at the AREF pin with a high impedant voltmeter.Note that VREF is a high impedant source, and only a capacitive load should be connected in a system.

If the user has a fixed voltage source connected to the AREF pin, the user may not use the other reference voltageoptions in the application, as they will be shorted to the external voltage. If no external voltage is applied to theAREF pin, the user may switch between AVCC and 2.56V as reference selection. The first ADC conversion resultafter switching reference voltage source may be inaccurate, and the user is advised to discard this result.

If differential channels are used, the selected reference should not be closer to AVCC than indicated in Table 27-4on page 303.

23.7 ADC Noise CancelerThe ADC features a noise canceler that enables conversion during sleep mode to reduce noise induced from theCPU core and other I/O peripherals. The noise canceler can be used with ADC Noise Reduction and Idle mode. Tomake use of this feature, the following procedure should be used:

1. Make sure that the ADC is enabled and is not busy converting. Single Conversion Mode must be selected and the ADC conversion complete interrupt must be enabled.

2. Enter ADC Noise Reduction mode (or Idle mode). The ADC will start a conversion once the CPU has been halted.

3. If no other interrupts occur before the ADC conversion completes, the ADC interrupt will wake up the CPU and execute the ADC Conversion Complete interrupt routine. If another interrupt wakes up the CPU before the ADC conversion is complete, that interrupt will be executed, and an ADC Conversion Complete interrupt request will be generated when the ADC conversion completes. The CPU will remain in active mode until a new sleep command is executed.

Note that the ADC will not be automatically turned off when entering other sleep modes than Idle mode and ADCNoise Reduction mode. The user is advised to write zero to ADEN before entering such sleep modes to avoidexcessive power consumption. If the ADC is enabled in such sleep modes and the user wants to perform differen-tial conversions, the user is advised to switch the ADC off and on after waking up from sleep to prompt anextended conversion to get a valid result.

23.7.1 Analog Input CircuitryThe Analog Input Circuitry for single ended channels is illustrated in Figure 22-8. An analog source applied toADCn is subjected to the pin capacitance and input leakage of that pin, regardless of whether that channel isselected as input for the ADC. When the channel is selected, the source must drive the S/H capacitor through theseries resistance (combined resistance in the input path).

The ADC is optimized for analog signals with an output impedance of approximately 10 k or less. If such a sourceis used, the sampling time will be negligible. If a source with higher impedance is used, the sampling time willdepend on how long time the source needs to charge the S/H capacitor, with can vary widely. The user is recom-mended to only use low impedant sources with slowly varying signals, since this minimizes the required chargetransfer to the S/H capacitor.

If differential gain channels are used, the input circuitry looks somewhat different, although source impedances of afew hundred k or less is recommended.

Signal components higher than the Nyquist frequency (fADC/2) should not be present for either kind of channels, toavoid distortion from unpredictable signal convolution. The user is advised to remove high frequency componentswith a low-pass filter before applying the signals as inputs to the ADC.


Figure 23-8. Analog Input Circuitry

23.7.2 Analog Noise Canceling TechniquesDigital circuitry inside and outside the device generates EMI which might affect the accuracy of analog measure-ments. If conversion accuracy is critical, the noise level can be reduced by applying the following techniques:

1. Keep analog signal paths as short as possible. Make sure analog tracks run over the analog ground plane, and keep them well away from high-speed switching digital tracks.

2. The AVCC pin on the device should be connected to the digital VCC supply voltage via an LC network as shown in Figure 22-9.

3. Use the ADC noise canceler function to reduce induced noise from the CPU.

4. If any ADC port pins are used as digital outputs, it is essential that these do not switch while a conversion is in progress.

Figure 23-9. ADC Power Connections

ADCn

IIH

1..100 kΩCS/H= 14 pF

VCC/2

IIL

GN

D

VC

C

PA

0 (A

DC

0)

PA

1 (A

DC

1)

PA

2 (A

DC

2)

PA

3 (A

DC

3)

PA4 (ADC4)

PA5 (ADC5)

PA6 (ADC6)

PA7 (ADC7)

AREF

AVCC

GND

PC7

10μH

100n

FA

nalo

g G

roun

d P

lane


23.7.3 Offset Compensation SchemesThe gain stage has a built-in offset cancellation circuitry that nulls the offset of differential measurements as muchas possible. The remaining offset in the analog path can be measured directly by selecting the same channel forboth differential inputs. This offset residue can be then subtracted in software from the measurement results. Usingthis kind of software based offset correction, offset on any channel can be reduced below one LSB.

23.7.4 ADC Accuracy DefinitionsAn n-bit single-ended ADC converts a voltage linearly between GND and VREF in 2n steps (LSBs). The lowest codeis read as 0, and the highest code is read as 2n-1.

Several parameters describe the deviation from the ideal behavior:

• Offset: The deviation of the first transition (0x000 to 0x001) compared to the ideal transition (at 0.5 LSB). Ideal value: 0 LSB.

Figure 23-10. Offset Error

• Gain Error: After adjusting for offset, the Gain Error is found as the deviation of the last transition (0x3FE to 0x3FF) compared to the ideal transition (at 1.5 LSB below maximum). Ideal value: 0 LSB

Figure 23-11. Gain Error

• Integral Non-linearity (INL): After adjusting for offset and gain error, the INL is the maximum deviation of an actual transition compared to an ideal transition for any code. Ideal value: 0 LSB.

Output Code

VREF Input Voltage

Ideal ADC

Actual ADC

OffsetError

Output Code

VREF Input Voltage

Ideal ADC

Actual ADC

GainError


Figure 23-12. Integral Non-linearity (INL)

• Differential Non-linearity (DNL): The maximum deviation of the actual code width (the interval between two adjacent transitions) from the ideal code width (1 LSB). Ideal value: 0 LSB.

Figure 23-13. Differential Non-linearity (DNL)

• Quantization Error: Due to the quantization of the input voltage into a finite number of codes, a range of input voltages (1 LSB wide) will code to the same value. Always ±0.5 LSB.

• Absolute Accuracy: The maximum deviation of an actual (unadjusted) transition compared to an ideal transition for any code. This is the compound effect of Offset, Gain Error, Differential Error, Non-linearity, and Quantization Error. Ideal value: ±0.5 LSB.

23.8 ADC Conversion ResultAfter the conversion is complete (ADIF is high), the conversion result can be found in the ADC Result Registers(ADCL, ADCH).

Output Code

VREF Input Voltage

Ideal ADC

Actual ADC

INL

Output Code

0x3FF

0x000

0 VREF Input Voltage

DNL

1 LSB


For single ended conversion, the result is

where VIN is the voltage on the selected input pin and VREF the selected voltage reference (see Table 22-3 on page222 and Table 22-4 on page 223). 0x000 represents analog ground, and 0x3FF represents the selected referencevoltage minus one LSB.

If differential channels are used, the result is

where VPOS is the voltage on the positive input pin, VNEG the voltage on the negative input pin, GAIN the selectedgain factor, and VREF the selected voltage reference. The result is presented in two’s complement form, from 0x200(-512d) through 0x1FF (+511d). Note that if the user wants to perform a quick polarity check of the results, it is suf-ficient to read the MSB of the result (ADC9 in ADCH). If this bit is one, the result is negative, and if this bit is zero,the result is positive. Figure 22-14 shows the decoding of the differential input range.

Table 22-2 shows the resulting output codes if the differential input channel pair (ADCn - ADCm) is selected with again of GAIN and a reference voltage of VREF.

Figure 23-14. Differential Measurement Range

ADCVIN 1024

VREF--------------------------=

ADCVPOS VNEG– GAIN 512

VREF------------------------------------------------------------------------=

0

Output Code

0x1FF

0x000

VREF/GAIN Differential InputVoltage (Volts)

0x3FF

0x200

- VREF/GAIN


Example:

ADMUX = 0xED (ADC3 - ADC2, 10x gain, 2.56V reference, left adjusted result)

Voltage on ADC3 is 300 mV, voltage on ADC2 is 500 mV.

ADCR = 512 × 10 × (300 - 500) / 2560 = -400 = 0x270

ADCL will thus read 0x00, and ADCH will read 0x9C. Writing zero to ADLAR right adjusts the result:ADCL = 0x70, ADCH = 0x02.

Table 23-2. Correlation between Input Voltage and Output Codes

VADCn Read code Corresponding Decimal Value

VADCm + VREF/GAIN 0x1FF 511

VADCm + 511/512 VREF/GAIN 0x1FF 511

VADCm + 510/512 VREF/GAIN 0x1FE 510

:. :. :.

VADCm + 1/512 VREF/GAIN 0x001 1

VADCm 0x000 0

VADCm - 1/512 VREF/GAIN 0x3FF -1

:. :. :.

VADCm - 511/512 VREF/GAIN 0x201 -511

VADCm - VREF/GAIN 0x200 -512



23.9.1 ADMUX – ADC Multiplexer Selection Register

• Bit 7:6 – REFS1:0: Reference Selection BitsThese bits select the voltage reference for the ADC, as shown in Table 22-3. If these bits are changed during aconversion, the change will not go in effect until this conversion is complete (ADIF in ADCSRA is set). The internalvoltage reference options may not be used if an external reference voltage is being applied to the AREF pin.

• Bit 5 – ADLAR: ADC Left Adjust ResultThe ADLAR bit affects the presentation of the ADC conversion result in the ADC Data Register. Write one toADLAR to left adjust the result. Otherwise, the result is right adjusted. Changing the ADLAR bit will affect the ADCData Register immediately, regardless of any ongoing conversions. For a complete description of this bit, see“ADCL and ADCH – The ADC Data Register” on page 225.

• Bits 4:0 – MUX4:0: Analog Channel and Gain Selection BitsThe value of these bits selects which combination of analog inputs are connected to the ADC. These bits alsoselect the gain for the differential channels. See Table 22-4 for details. If these bits are changed during aconversion, the change will not go in effect until this conversion is complete (ADIF in ADCSRA is set).

Bit 7 6 5 4 3 2 1 0

REFS1 REFS0 ADLAR MUX4 MUX3 MUX2 MUX1 MUX0 ADMUX



Table 23-3. Voltage Reference Selections for ADC

REFS1 REFS0 Voltage Reference Selection

0 0 AREF, Internal Vref turned off

0 1 AVCC with external capacitor at AREF pin

1 0 Reserved

1 1 Internal 2.56V Voltage Reference with external capacitor at AREF pin

Table 23-4. Input Channel and Gain Selections

MUX4:0Single Ended Input

Positive Differential Input

Negative Differential Input Gain

00000 ADC0

00001 ADC1

00010 ADC2

00011 ADC3 N/A

00100 ADC4

00101 ADC5

00110 ADC6

00111 ADC7

01000 ADC0 ADC0 10x

01001 ADC1 ADC0 10x

01010 ADC0 ADC0 200x

01011 ADC1 ADC0 200x


23.9.2 ADCSRA – ADC Control and Status Register A

• Bit 7 – ADEN: ADC EnableWriting this bit to one enables the ADC. By writing it to zero, the ADC is turned off. Turning the ADC off while a con-version is in progress, will terminate this conversion.

• Bit 6 – ADSC: ADC Start ConversionIn Single Conversion mode, write this bit to one to start each conversion. In Free Running Mode, write this bit toone to start the first conversion. The first conversion after ADSC has been written after the ADC has been enabled,or if ADSC is written at the same time as the ADC is enabled, will take 25 ADC clock cycles instead of the normal13. This first conversion performs initialization of the ADC.

ADSC will read as one as long as a conversion is in progress. When the conversion is complete, it returns to zero.Writing zero to this bit has no effect.

01100 ADC2 ADC2 10x

01101 ADC3 ADC2 10x

01110 ADC2 ADC2 200x

01111 ADC3 ADC2 200x

10000 ADC0 ADC1 1x

10001 ADC1 ADC1 1x

10010 N/A ADC2 ADC1 1x

10011 ADC3 ADC1 1x

10100 ADC4 ADC1 1x

10101 ADC5 ADC1 1x

10110 ADC6 ADC1 1x

10111 ADC7 ADC1 1x

11000 ADC0 ADC2 1x

11001 ADC1 ADC2 1x

11010 ADC2 ADC2 1x

11011 ADC3 ADC2 1x

11100 ADC4 ADC2 1x

11101 ADC5 ADC2 1x

11110 1.22V (VBG) N/A

11111 0V (GND)

Table 23-4. Input Channel and Gain Selections (Continued)

MUX4:0Single Ended Input

Positive Differential Input

Negative Differential Input Gain

Bit 7 6 5 4 3 2 1 0

ADEN ADSC ADATE ADIF ADIE ADPS2 ADPS1 ADPS0 ADCSRA




• Bit 5 – ADATE: ADC Auto Trigger EnableWhen this bit is written to one, Auto Triggering of the ADC is enabled. The ADC will start a conversion on a positiveedge of the selected trigger signal. The trigger source is selected by setting the ADC Trigger Select bits, ADTS inSFIOR.

• Bit 4 – ADIF: ADC Interrupt FlagThis bit is set when an ADC conversion completes and the Data Registers are updated. The ADC ConversionComplete Interrupt is executed if the ADIE bit and the I-bit in SREG are set. ADIF is cleared by hardware whenexecuting the corresponding interrupt handling vector. Alternatively, ADIF is cleared by writing a logical one to theflag. Beware that if doing a Read-Modify-Write on ADCSRA, a pending interrupt can be disabled. This also appliesif the SBI and CBI instructions are used.

• Bit 3 – ADIE: ADC Interrupt EnableWhen this bit is written to one and the I-bit in SREG is set, the ADC Conversion Complete Interrupt is activated.

• Bits 2:0 – ADPS2:0: ADC Prescaler Select BitsThese bits determine the division factor between the XTAL frequency and the input clock to the ADC.

23.9.3 ADCL and ADCH – The ADC Data Register

23.9.3.1 ADLAR = 0

Table 23-5. ADC Prescaler Selections

ADPS2 ADPS1 ADPS0 Division Factor

0 0 0 2

0 0 1 2

0 1 0 4

0 1 1 8

1 0 0 16

1 0 1 32

1 1 0 64

1 1 1 128

Bit 15 14 13 12 11 10 9 8

– – – – – – ADC9 ADC8 ADCH

ADC7 ADC6 ADC5 ADC4 ADC3 ADC2 ADC1 ADC0 ADCL

7 6 5 4 3 2 1 0


R R R R R R R R


0 0 0 0 0 0 0 0


23.9.3.2 ADLAR = 1

When an ADC conversion is complete, the result is found in these two registers. If differential channels are used,the result is presented in two’s complement form.

When ADCL is read, the ADC Data Register is not updated until ADCH is read. Consequently, if the result is leftadjusted and no more than 8-bit precision is required, it is sufficient to read ADCH. Otherwise, ADCL must be readfirst, then ADCH.

The ADLAR bit in ADMUX, and the MUXn bits in ADMUX affect the way the result is read from the registers. IfADLAR is set, the result is left adjusted. If ADLAR is cleared (default), the result is right adjusted.

• ADC9:0: ADC Conversion ResultThese bits represent the result from the conversion, as detailed in “ADC Conversion Result” on page 220.

23.9.4 SFIOR – Special FunctionIO Register

• Bit 7:5 – ADTS2:0: ADC Auto Trigger SourceIf ADATE in ADCSRA is written to one, the value of these bits selects which source will trigger an ADC conversion.If ADATE is cleared, the ADTS2:0 settings will have no effect. A conversion will be triggered by the rising edge ofthe selected Interrupt Flag. Note that switching from a trigger source that is cleared to a trigger source that is set,will generate a positive edge on the trigger signal. If ADEN in ADCSRA is set, this will start a conversion. Switchingto Free Running mode (ADTS[2:0]=0) will not cause a trigger event, even if the ADC Interrupt Flag is set.

• Bit 4 – Reserved BitThis bit is reserved for future use in the ATmega32A. For ensuring compability with future devices, this bit must bewritten zero when SFIOR is written.

Bit 15 14 13 12 11 10 9 8

ADC9 ADC8 ADC7 ADC6 ADC5 ADC4 ADC3 ADC2 ADCH

ADC1 ADC0 – – – – – – ADCL

7 6 5 4 3 2 1 0


R R R R R R R R


0 0 0 0 0 0 0 0

Bit 7 6 5 4 3 2 1 0




Table 23-6. ADC Auto Trigger Source Selections

ADTS2 ADTS1 ADTS0 Trigger Source

0 0 0 Free Running mode

0 0 1 Analog Comparator

0 1 0 External Interrupt Request 0

0 1 1 Timer/Counter0 Compare Match

1 0 0 Timer/Counter0 Overflow

1 0 1 Timer/Counter1 Compare Match B

1 1 0 Timer/Counter1 Overflow

1 1 1 Timer/Counter1 Capture Event


24. JTAG Interface and On-chip Debug System

24.1 Features• JTAG (IEEE std. 1149.1 Compliant) Interface• Boundary-scan Capabilities According to the IEEE std. 1149.1 (JTAG) Standard• Debugger Access to:

– All Internal Peripheral Units– Internal and External RAM– The Internal Register File– Program Counter– EEPROM and Flash Memories– Extensive On-chip Debug Support for Break Conditions, Including– AVR Break Instruction– Break on Change of Program Memory Flow– Single Step Break– Program Memory Breakpoints on Single Address or Address Range– Data Memory Breakpoints on Single Address or Address Range

• Programming of Flash, EEPROM, Fuses, and Lock Bits through the JTAG Interface• On-chip Debugging Supported by AVR Studio®

24.2 OverviewThe AVR IEEE std. 1149.1 compliant JTAG interface can be used for

• Testing PCBs by using the JTAG Boundary-scan capability

• Programming the non-volatile memories, Fuses and Lock bits

• On-chip Debugging

A brief description is given in the following sections. Detailed descriptions for Programming via the JTAG interface,and using the Boundary-scan Chain can be found in the sections “Programming via the JTAG Interface” on page284 and “IEEE 1149.1 (JTAG) Boundary-scan” on page 233, respectively. The On-chip Debug support is consid-ered being private JTAG instructions, and distributed within ATMEL and to selected third party vendors only.

Figure 23-1 shows a block diagram of the JTAG interface and the On-chip Debug system. The TAP Controller is astate machine controlled by the TCK and TMS signals. The TAP Controller selects either the JTAG InstructionRegister or one of several Data Registers as the scan chain (Shift Register) between the TDI input and TDO out-put. The Instruction Register holds JTAG instructions controlling the behavior of a Data Register.

The ID-Register, Bypass Register, and the Boundary-scan Chain are the Data Registers used for board-level test-ing. The JTAG Programming Interface (actually consisting of several physical and virtual Data Registers) is usedfor JTAG Serial Programming via the JTAG interface. The Internal Scan Chain and Break Point Scan Chain areused for On-chip Debugging only.

24.3 TAP – Test Access PortThe JTAG interface is accessed through four of the AVR’s pins. In JTAG terminology, these pins constitute theTest Access Port – TAP. These pins are:

• TMS: Test Mode Select. This pin is used for navigating through the TAP-controller state machine.

• TCK: Test Clock. JTAG operation is synchronous to TCK.

• TDI: Test Data In. Serial input data to be shifted in to the Instruction Register or Data Register (Scan Chains).


• TDO: Test Data Out. Serial output data from Instruction Register or Data Register.

The IEEE std. 1149.1 also specifies an optional TAP signal; TRST – Test ReSeT – which is not provided.

When the JTAGEN fuse is unprogrammed, these four TAP pins are normal port pins and the TAP controller is inreset. When programmed and the JTD bit in MCUCSR is cleared, the TAP input signals are internally pulled highand the JTAG is enabled for Boundary-scan and programming. In this case, the TAP output pin (TDO) is left float-ing in states where the JTAG TAP controller is not shifting data, and must therefore be connected to a pull-upresistor or other hardware having pull-ups (for instance the TDI-input of the next device in the scan chain). Thedevice is shipped with this fuse programmed.

For the On-chip Debug system, in addition to the JTAG interface pins, the RESET pin is monitored by the debug-ger to be able to detect external reset sources. The debuggerbta can also pull the RESET pin low to reset thewhole system, assuming only open collectors on the reset line are used in the application.

Figure 24-1. Block Diagram

TAPCONTROLLER

TDITDOTCKTMS

FLASHMEMORY

AVR CPU

DIGITALPERIPHERAL

UNITS

JTAG / AVR CORECOMMUNICATION

INTERFACE

BREAKPOINTUNIT

FLOW CONTROLUNIT

OCD STATUSAND CONTROL

INTERNAL SCANCHAIN

MUX

INSTRUCTIONREGISTER

IDREGISTER

BYPASSREGISTER

JTAG PROGRAMMINGINTERFACE

PCInstruction

AddressData

BREAKPOINTSCAN CHAIN

ADDRESSDECODER

AN

AL

OG

PE

RIP

HE

RIA

LU

NIT

S

I/O PORT 0

I/O PORT n

BOUNDARY SCAN CHAIN

An

alo

g in

pu

tsC

on

tro

l & C

lock

lin

es

DEVICE BOUNDARY


Figure 24-2. TAP Controller State Diagram

24.4 TAP ControllerThe TAP controller is a 16-state finite state machine that controls the operation of the Boundary-scan circuitry,JTAG programming circuitry, or On-chip Debug system. The state transitions depicted in Figure 23-2 depend onthe signal present on TMS (shown adjacent to each state transition) at the time of the rising edge at TCK. The ini-tial state after a Power-On Reset is Test-Logic-Reset.

As a definition in this document, the LSB is shifted in and out first for all Shift Registers.

Assuming Run-Test/Idle is the present state, a typical scenario for using the JTAG interface is:

• At the TMS input, apply the sequence 1, 1, 0, 0 at the rising edges of TCK to enter the Shift Instruction Register – Shift-IR state. While in this state, shift the four bits of the JTAG instructions into the JTAG Instruction Register from the TDI input at the rising edge of TCK. The TMS input must be held low during input of the 3 LSBs in order to remain in the Shift-IR state. The MSB of the instruction is shifted in when this state is left by setting TMS high. While the instruction is shifted in from the TDI pin, the captured IR-state 0x01 is shifted out on the

Test-Logic-Reset

Run-Test/Idle

Shift-DR

Exit1-DR

Pause-DR

Exit2-DR

Update-DR

Select-IR Scan

Capture-IR

Shift-IR

Exit1-IR

Pause-IR

Exit2-IR

Update-IR

Select-DR Scan

Capture-DR

0

1

0 1 1 1

0 0

0 0

1 1

1 0

1

1

0

1

0

0

1 0

1

1

0

1

0

0

00

11


TDO pin. The JTAG Instruction selects a particular Data Register as path between TDI and TDO and controls the circuitry surrounding the selected Data Register.

• Apply the TMS sequence 1, 1, 0 to re-enter the Run-Test/Idle state. The instruction is latched onto the parallel output from the Shift Register path in the Update-IR state. The Exit-IR, Pause-IR, and Exit2-IR states are only used for navigating the state machine.

• At the TMS input, apply the sequence 1, 0, 0 at the rising edges of TCK to enter the Shift Data Register – Shift-DR state. While in this state, upload the selected Data Register (selected by the present JTAG instruction in the JTAG Instruction Register) from the TDI input at the rising edge of TCK. In order to remain in the Shift-DR state, the TMS input must be held low during input of all bits except the MSB. The MSB of the data is shifted in when this state is left by setting TMS high. While the Data Register is shifted in from the TDI pin, the parallel inputs to the Data Register captured in the Capture-DR state is shifted out on the TDO pin.

• Apply the TMS sequence 1, 1, 0 to re-enter the Run-Test/Idle state. If the selected Data Register has a latched parallel-output, the latching takes place in the Update-DR state. The Exit-DR, Pause-DR, and Exit2-DR states are only used for navigating the state machine.

As shown in the state diagram, the Run-Test/Idle state need not be entered between selecting JTAG instructionand using Data Registers, and some JTAG instructions may select certain functions to be performed in the Run-Test/Idle, making it unsuitable as an Idle state.

Note: Independent of the initial state of the TAP Controller, the Test-Logic-Reset state can always be entered by holding TMS high for five TCK clock periods.

For detailed information on the JTAG specification, refer to the literature listed in “Bibliography” on page 232.

24.5 Using the Boundary-scan ChainA complete description of the Boundary-scan capabilities are given in the section “IEEE 1149.1 (JTAG) Boundary-scan” on page 233.

24.6 Using the On-chip Debug SystemAs shown in Figure 23-1, the hardware support for On-chip Debugging consists mainly of:

• A scan chain on the interface between the internal AVR CPU and the internal peripheral units

• Break Point unit

• Communication interface between the CPU and JTAG system

All read or modify/write operations needed for implementing the Debugger are done by applying AVR instructionsvia the internal AVR CPU Scan Chain. The CPU sends the result to an I/O memory mapped location which is partof the communication interface between the CPU and the JTAG system.

The Break Point Unit implements Break on Change of Program Flow, Single Step Break, 2 Program Memory BreakPoints, and 2 combined Break Points. Together, the 4 Break Points can be configured as either:

• 4 single Program Memory Break Points

• 3 Single Program Memory Break Point + 1 single Data Memory Break Point

• 2 single Program Memory Break Points + 2 single Data Memory Break Points

• 2 single Program Memory Break Points + 1 Program Memory Break Point with mask (“range Break Point”)

• 2 single Program Memory Break Points + 1 Data Memory Break Point with mask (“range Break Point”)

A debugger, like the AVR Studio, may however use one or more of these resources for its internal purpose, leavingless flexibility to the end-user.

A list of the On-chip Debug specific JTAG instructions is given in “On-chip Debug Specific JTAG Instructions” onpage 231.


The JTAGEN Fuse must be programmed to enable the JTAG Test Access Port. In addition, the OCDEN Fuse mustbe programmed and no Lock bits must be set for the On-chip Debug system to work. As a security feature, the On-chip Debug system is disabled when any Lock bits are set. Otherwise, the On-chip Debug system would have pro-vided a back-door into a secured device.

The AVR JTAG ICE from Atmel is a powerful development tool for On-chip Debugging of all AVR 8-bit RISC Micro-controllers with IEEE 1149.1 compliant JTAG interface. The JTAG ICE and the AVR Studio user interface give theuser complete control of the internal resources of the microcontroller, helping to reduce development time by mak-ing debugging easier. The JTAG ICE performs real-time emulation of the micrcontroller while it is running in atarget system.

Please refer to the Support Tools section on the AVR pages on www.atmel.com for a full description of the AVRJTEG ICE. AVR Studio can be downloaded free from Software section on the same web site.

All necessary execution commands are available in AVR Studio, both on source level and on disassembly level.The user can execute the program, single step through the code either by tracing into or stepping over functions,step out of functions, place the cursor on a statement and execute until the statement is reached, stop the execu-tion, and reset the execution target. In addition, the user can have an unlimited number of code breakpoints (usingthe BREAK instruction) and up to two data memory breakpoints, alternatively combined as a mask (range) BreakPoint.

24.7 On-chip Debug Specific JTAG InstructionsThe On-chip Debug support is considered being private JTAG instructions, and distributed within ATMEL and toselected third party vendors only. Instruction opcodes are listed for reference.

24.7.1 PRIVATE0; $8Private JTAG instruction for accessing On-chip Debug system.

24.7.2 PRIVATE1; $9Private JTAG instruction for accessing On-chip Debug system.

24.7.3 PRIVATE2; $APrivate JTAG instruction for accessing On-chip Debug system.

24.7.4 PRIVATE3; $BPrivate JTAG instruction for accessing On-chip Debug system.

24.8 Using the JTAG Programming CapabilitiesProgramming of AVR parts via JTAG is performed via the 4-pin JTAG port, TCK, TMS, TDI and TDO. These arethe only pins that need to be controlled/observed to perform JTAG programming (in addition to power pins). It is notrequired to apply 12V externally. The JTAGEN fuse must be programmed and the JTD bit in the MCUSR Registermust be cleared to enable the JTAG Test Access Port.

The JTAG programming capability supports:

• Flash programming and verifying

• EEPROM programming and verifying

• Fuse programming and verifying

• Lock bit programming and verifying

The Lock bit security is exactly as in Parallel Programming mode. If the Lock bits LB1 or LB2 are programmed, theOCDEN Fuse cannot be programmed unless first doing a chip erase. This is a security feature that ensures noback-door exists for reading out the content of a secured device.


The details on programming through the JTAG interface and programming specific JTAG instructions are given inthe section “Programming via the JTAG Interface” on page 284.


24.9.1 OCDR – On-chip Debug Register

The OCDR Register provides a communication channel from the running program in the microcontroller to thedebugger. The CPU can transfer a byte to the debugger by writing to this location. At the same time, an InternalFlag; I/O Debug Register Dirty – IDRD – is set to indicate to the debugger that the register has been written. Whenthe CPU reads the OCDR Register the 7 LSB will be from the OCDR Register, while the MSB is the IDRD bit. Thedebugger clears the IDRD bit when it has read the information.

In some AVR devices, this register is shared with a standard I/O location. In this case, the OCDR Register can onlybe accessed if the OCDEN Fuse is programmed, and the debugger enables access to the OCDR Register. In allother cases, the standard I/O location is accessed.

Refer to the debugger documentation for further information on how to use this register.

24.10 BibliographyFor more information about general Boundary-scan, the following literature can be consulted:

• IEEE: IEEE Std 1149.1-1990. IEEE Standard Test Access Port and Boundary-scan Architecture, IEEE, 1993

• Colin Maunder: The Board Designers Guide to Testable Logic Circuits, Addison-Wesley, 1992

Bit 7 6 5 4 3 2 1 0

MSB/IDRD LSB OCDR




25. IEEE 1149.1 (JTAG) Boundary-scan

25.1 Features• JTAG (IEEE std. 1149.1 Compliant) Interface• Boundary-scan Capabilities According to the JTAG Standard• Full Scan of all Port Functions as well as Analog Circuitry having Off-chip Connections• Supports the Optional IDCODE Instruction• Additional Public AVR_RESET Instruction to Reset the AVR

25.2 OverviewThe Boundary-scan chain has the capability of driving and observing the logic levels on the digital I/O pins, as wellas the boundary between digital and analog logic for analog circuitry having Off-chip connections. At system level,all ICs having JTAG capabilities are connected serially by the TDI/TDO signals to form a long Shift Register. Anexternal controller sets up the devices to drive values at their output pins, and observe the input values receivedfrom other devices. The controller compares the received data with the expected result. In this way, Boundary-scanprovides a mechanism for testing interconnections and integrity of components on Printed Circuits Boards by usingthe four TAP signals only.

The four IEEE 1149.1 defined mandatory JTAG instructions IDCODE, BYPASS, SAMPLE/PRELOAD, andEXTEST, as well as the AVR specific public JTAG instruction AVR_RESET can be used for testing the Printed Cir-cuit Board. Initial scanning of the Data Register path will show the ID-code of the device, since IDCODE is thedefault JTAG instruction. It may be desirable to have the AVR device in Reset during Test mode. If not reset, inputsto the device may be determined by the scan operations, and the internal software may be in an undeterminedstate when exiting the Test mode. Entering reset, the outputs of any Port Pin will instantly enter the high imped-ance state, making the HIGHZ instruction redundant. If needed, the BYPASS instruction can be issued to make theshortest possible scan chain through the device. The device can be set in the reset state either by pulling the exter-nal RESET pin low, or issuing the AVR_RESET instruction with appropriate setting of the Reset Data Register.

The EXTEST instruction is used for sampling external pins and loading output pins with data. The data from theoutput latch will be driven out on the pins as soon as the EXTEST instruction is loaded into the JTAG IR-Register.Therefore, the SAMPLE/PRELOAD should also be used for setting initial values to the scan ring, to avoid damag-ing the board when issuing the EXTEST instruction for the first time. SAMPLE/PRELOAD can also be used fortaking a snapshot of the external pins during normal operation of the part.

The JTAGEN Fuse must be programmed and the JTD bit in the I/O Register MCUCSR must be cleared to enablethe JTAG Test Access Port.

When using the JTAG interface for Boundary-scan, using a JTAG TCK clock frequency higher than the internalchip frequency is possible. The chip clock is not required to run.

25.3 Data Registers The Data Registers relevant for Boundary-scan operations are:

• Bypass Register

• Device Identification Register

• Reset Register

• Boundary-scan Chain


25.3.1 Bypass RegisterThe Bypass Register consists of a single Shift Register stage. When the Bypass Register is selected as pathbetween TDI and TDO, the register is reset to 0 when leaving the Capture-DR controller state. The Bypass Regis-ter can be used to shorten the scan chain on a system when the other devices are to be tested.

25.3.2 Device Identification RegisterFigure 24-1 shows the structure of the Device Identification Register.

Figure 25-1. The Format of the Device Identification Register

25.3.2.1 VersionVersion is a 4-bit number identifying the revision of the component. The JTAG version number follows the revisionof the device. Revision A is 0x0, revision B is x1 and so on.

25.3.2.2 Part NumberThe part number is a 16-bit code identifying the component. The JTAG Part Number for ATmega32A is listed inTable 24-1.

25.3.2.3 Manufacturer IDThe Manufacturer ID is a 11 bit code identifying the manufacturer. The JTAG manufacturer ID for ATMEL is listedin Table 24-2.

25.3.3 Reset RegisterThe Reset Register is a Test Data Register used to reset the part. Since the AVR tri-states Port Pins when reset,the Reset Register can also replace the function of the unimplemented optional JTAG instruction HIGHZ.

A high value in the Reset Register corresponds to pulling the External Reset low. The part is reset as long as thereis a high value present in the Reset Register. Depending on the Fuse settings for the clock options, the part willremain reset for a Reset Time-Out Period (refer to “Clock Sources” on page 26) after releasing the Reset Register.The output from this Data Register is not latched, so the reset will take place immediately, as shown in Figure 24-2.

MSB LSB

Bit 31 28 27 12 11 1 0

Device ID Version Part Number Manufacturer ID 1

4 bits 16 bits 11 bits 1 bit

Table 25-1. AVR JTAG Part Number

Part Number JTAG Part Number (Hex)

ATmega32A 0x9502

Table 25-2. Manufacturer ID

Manufacturer JTAG Man. ID (Hex)

ATMEL 0x01F


Figure 25-2. Reset Register

25.3.4 Boundary-scan ChainThe Boundary-scan Chain has the capability of driving and observing the logic levels on the digital I/O pins, as wellas the boundary between digital and analog logic for analog circuitry having Off-chip connections.

See “Boundary-scan Chain” on page 236 for a complete description.

25.4 Boundary-scan Specific JTAG InstructionsThe instruction register is 4-bit wide, supporting up to 16 instructions. Listed below are the JTAG instructions usefulfor Boundary-scan operation. Note that the optional HIGHZ instruction is not implemented, but all outputs with tri-state capability can be set in high-impedance state by using the AVR_RESET instruction, since the initial state forall port pins is tri-state.

As a definition in this datasheet, the LSB is shifted in and out first for all Shift Registers.

The OPCODE for each instruction is shown behind the instruction name in hex format. The text describes whichData Register is selected as path between TDI and TDO for each instruction.

25.4.1 EXTEST; $0Mandatory JTAG instruction for selecting the Boundary-scan Chain as Data Register for testing circuitry external tothe AVR package. For port-pins, Pull-up Disable, Output Control, Output Data, and Input Data are all accessible inthe scan chain. For Analog circuits having Off-chip connections, the interface between the analog and the digitallogic is in the scan chain. The contents of the latched outputs of the Boundary-scan chain is driven out as soon asthe JTAG IR-register is loaded with the EXTEST instruction.

The active states are:

• Capture-DR: Data on the external pins are sampled into the Boundary-scan Chain.

• Shift-DR: The Internal Scan Chain is shifted by the TCK input.

• Update-DR: Data from the scan chain is applied to output pins.

25.4.2 IDCODE; $1Optional JTAG instruction selecting the 32-bit ID-register as Data Register. The ID-register consists of a versionnumber, a device number and the manufacturer code chosen by JEDEC. This is the default instruction after power-up.


• Capture-DR: Data in the IDCODE-register is sampled into the Boundary-scan Chain.

• Shift-DR: The IDCODE scan chain is shifted by the TCK input.

D QFromTDI

ClockDR · AVR_RESET

To TDO

From other Internal andExternal Reset Sources

Internal Reset


25.4.3 SAMPLE_PRELOAD; $2Mandatory JTAG instruction for pre-loading the output latches and talking a snap-shot of the input/output pins with-out affecting the system operation. However, the output latches are not connected to the pins. The Boundary-scanChain is selected as Data Register.


• Capture-DR: Data on the external pins are sampled into the Boundary-scan Chain.

• Shift-DR: The Boundary-scan Chain is shifted by the TCK input.

• Update-DR: Data from the Boundary-scan Chain is applied to the output latches. However, the output latches are not connected to the pins.

25.4.4 AVR_RESET; $CThe AVR specific public JTAG instruction for forcing the AVR device into the Reset mode or releasing the JTAGReset source. The TAP controller is not reset by this instruction. The one bit Reset Register is selected as DataRegister. Note that the reset will be active as long as there is a logic 'one' in the Reset Chain. The output from thischain is not latched.


• Shift-DR: The Reset Register is shifted by the TCK input.

25.4.5 BYPASS; $FMandatory JTAG instruction selecting the Bypass Register for Data Register.


• Capture-DR: Loads a logic “0” into the Bypass Register.

• Shift-DR: The Bypass Register cell between TDI and TDO is shifted.

25.5 Boundary-scan ChainThe Boundary-scan chain has the capability of driving and observing the logic levels on the digital I/O pins, as wellas the boundary between digital and analog logic for analog circuitry having Off-chip connection.

25.5.1 Scanning the Digital Port PinsFigure 24-3 shows the Boundary-scan Cell for a bi-directional port pin with pull-up function. The cell consists of astandard Boundary-scan cell for the Pull-up Enable – PUExn – function, and a bi-directional pin cell that combinesthe three signals Output Control – OCxn, Output Data – ODxn, and Input Data – IDxn, into only a two-stage ShiftRegister. The port and pin indexes are not used in the following description.

The Boundary-scan logic is not included in the figures in the datasheet. Figure 24-4 shows a simple digital Port Pinas described in the section “I/O Ports” on page 50. The Boundary-scan details from Figure 24-3 replaces thedashed box in Figure 24-4.

When no alternate port function is present, the Input Data – ID – corresponds to the PINxn Register value (but IDhas no synchronizer), Output Data corresponds to the PORT Register, Output Control corresponds to the DataDirection – DD Register, and the Pull-up Enable – PUExn – corresponds to logic expression PUD · DDxn ·PORTxn.

Digital alternate port functions are connected outside the dotted box in Figure 24-4 to make the scan chain read theactual pin value. For Analog function, there is a direct connection from the external pin to the analog circuit, and ascan chain is inserted on the interface between the digital logic and the analog circuitry.


Figure 25-3. Boundary-scan Cell for Bidirectional Port Pin with Pull-up Function.

D Q D Q

G

0

10

1

D Q D Q

G

0

10

1

0

1

0

1D Q D Q

G

0

1

Port Pin (PXn)

VccEXTESTTo Next CellShiftDR

Output Control (OC)

Pullup Enable (PUE)

Output Data (OD)

Input Data (ID)

From Last Cell UpdateDRClockDR

FF2 LD2

FF1 LD1

LD0FF0


Figure 25-4. General Port Pin Schematic Diagram(1)

Note: 1. See Boundary-scan description for details.

25.5.2 Boundary-scan and the Two-wire InterfaceThe 2 Two-wire Interface pins SCL and SDA have one additional control signal in the scan-chain; Two-wire Inter-face Enable – TWIEN. As shown in Figure 24-5, the TWIEN signal enables a tri-state buffer with slew-rate controlin parallel with the ordinary digital port pins. A general scan cell as shown in Figure 24-9 is attached to the TWIENsignal.

Notes: 1. A separate scan chain for the 50 ns spike filter on the input is not provided. The ordinary scan support for digital port pins suffice for connectivity tests. The only reason for having TWIEN in the scan path, is to be able to discon-nect the slew-rate control buffer when doing boundary-scan.

2. Make sure the OC and TWIEN signals are not asserted simultaneously, as this will lead to drive contention.

CLK

RPx

RRx

WPx

RDx

WDx

PUD

SYNCHRONIZER

WDx: WRITE DDRx

WPx: WRITE PORTxRRx: READ PORTx REGISTERRPx: READ PORTx PIN

PUD: PULLUP DISABLE

CLK : I/O CLOCK

RDx: READ DDRx

D

L

Q

Q

RESET

RESET

Q

QD

Q

Q D

CLR

PORTxn

Q

Q D

CLR

DDxn

PINxn

DAT

A B

US

SLEEP

SLEEP: SLEEP CONTROL

Pxn

I/O

I/O

PUExn

OCxn

ODxn

IDxn

PUExn: PULLUP ENABLE for pin PxnOCxn: OUTPUT CONTROL for pin PxnODxn: OUTPUT DATA to pin PxnIDxn: INPUT DATA from pin Pxn


Figure 25-5. Additional Scan Signal for the Two-wire Interface

25.5.3 Scanning the RESET PinThe RESET pin accepts 5V active low logic for standard reset operation, and 12V active high logic for High VoltageParallel Programming. An observe-only cell as shown in Figure 24-6 is inserted both for the 5V reset signal; RSTT,and the 12V reset signal; RSTHV.

Figure 25-6. Observe-only Cell

25.5.4 Scanning the Clock PinsThe AVR devices have many clock options selectable by fuses. These are: Internal RC Oscillator, External RC,External Clock, (High Frequency) Crystal Oscillator, Low Frequency Crystal Oscillator, and Ceramic Resonator.

Figure 24-7 shows how each Oscillator with external connection is supported in the scan chain. The Enable signalis supported with a general boundary-scan cell, while the Oscillator/Clock output is attached to an observe-onlycell. In addition to the main clock, the Timer Oscillator is scanned in the same way. The output from the internal RCOscillator is not scanned, as this Oscillator does not have external connections.

Pxn

PUExn

ODxn

IDxn

TWIEN

OCxn

Slew-rate Limited

SRC

0

1D Q

FromPrevious

Cell

ClockDR

ShiftDR

ToNextCell

From System Pin To System Logic

FF1


Figure 25-7. Boundary-scan Cells for Oscillators and Clock Options

Table 24-3 summaries the scan registers for the external clock pin XTAL1, Oscillators with XTAL1/XTAL2 connec-tions as well as 32kHz Timer Oscillator.

Notes: 1. Do not enable more than one clock source as main clock at a time.

2. Scanning an Oscillator output gives unpredictable results as there is a frequency drift between the Internal Oscilla-tor and the JTAG TCK clock. If possible, scanning an external clock is preferred.

3. The clock configuration is programmed by fuses. As a fuse is not changed run-time, the clock configuration is con-sidered fixed for a given application. The user is advised to scan the same clock option as to be used in the final system. The enable signals are supported in the scan chain because the system logic can disable clock options in sleep modes, thereby disconnecting the Oscillator pins from the scan path if not provided. The INTCAP fuses are not supported in the scan-chain, so the boundary scan chain can not make a XTAL Oscillator requiring internal capacitors to run unless the fuse is correctly programmed.

25.5.5 Scanning the Analog ComparatorThe relevant Comparator signals regarding Boundary-scan are shown in Figure 24-8. The Boundary-scan cell fromFigure 24-9 is attached to each of these signals. The signals are described in Table 24-4.

The Comparator need not be used for pure connectivity testing, since all analog inputs are shared with a digitalport pin as well.

Table 25-3. Scan Signals for the Oscillators(1)(2)(3)

Enable Signal Scanned Clock Line Clock Option Scanned Clock Line when not Used

EXTCLKEN EXTCLK (XTAL1) External Clock 0

OSCON OSCCK External CrystalExternal Ceramic Resonator

0

RCOSCEN RCCK External RC 1

OSC32EN OSC32CK Low Freq. External Crystal 0

TOSKON TOSCK 32kHz Timer Oscillator 0

0

1D Q

FromPrevious

Cell

ClockDR

ShiftDR

ToNextCell

To

Sys

tem

Log

ic

FF10

1D Q D Q

G

0

1

FromPrevious

Cell

ClockDR UpdateDR

ShiftDR

ToNextCell EXTEST

Fro

m D

igita

l Log

ic

XTAL1/TOSC1 XTAL2/TOSC2

Oscillator

ENABLE OUTPUT


Figure 25-8. Analog Comparator

Figure 25-9. General Boundary-scan Cell used for Signals for Comparator and ADC

ACBG

BANDGAPREFERENCE


ACME

AC_IDLE

ACO

ADCEN

0

1D Q D Q

G

0

1

FromPrevious

Cell

ClockDR UpdateDR

ShiftDR

ToNextCell EXTEST

To Analog Circuitry/To Digital Logic

From Digital Logic/From Analog Ciruitry


25.5.6 Scanning the ADC

Figure 24-10 shows a block diagram of the ADC with all relevant control and observe signals. The Boundary-scan cell fromFigure 24-9 is attached to each of these signals. The ADC need not be used for pure connectivity testing, since all analoginputs are shared with a digital port pin as well.

Figure 25-10. Analog to Digital Converter

The signals are described briefly in Table 24-5.

Table 25-4. Boundary-scan Signals for the Analog Comparator

Signal Name

Direction as Seen from the Comparator Description

Recommended Input when Not in Use

Output Values when Recommended Inputs are Used

AC_IDLE Input Turns off Analogcomparator when true

1 Depends upon µC code being executed

ACO Output Analog ComparatorOutput

Will become input to µC code being executed

0

ACME Input Uses output signal fromADC mux when true


ACBG Input Bandgap Referenceenable


10-bit DAC +

-

AREF

PRECH

DACOUT

COMP

MUXEN_7ADC_7

MUXEN_6ADC_6

MUXEN_5ADC_5

MUXEN_4ADC_4

MUXEN_3ADC_3

MUXEN_2ADC_2

MUXEN_1ADC_1

MUXEN_0ADC_0

NEGSEL_2ADC_2

NEGSEL_1ADC_1

NEGSEL_0ADC_0

EXTCH

+

-

+

-10x 20x

G10 G20

STACLK

AMPEN

2.56Vref

IREFEN

AREF

VCCREN

DAC_9..0

ADCEN

HOLD

PRECH

GNDEN

PASSEN

ACTEN

CO

MP

SCTESTADCBGEN

To Comparator

1.22Vref AREF


Table 25-5. Boundary-scan Signals for the ADC

Signal Name

Direction as Seenfrom the ADC Description


Output Values when Recommended Inputs are used, and CPU is not Using the ADC

COMP Output Comparator Output 0 0

ACLK Input Clock signal to gain stages implemented as Switch-cap filters

0 0

ACTEN Input Enable path from gain stages to the comparator

0 0

ADCBGEN Input Enable Band-gap reference as negative input to comparator

0 0

ADCEN Input Power-on signal to the ADC 0 0

AMPEN Input Power-on signal to the gain stages 0 0

DAC_9 Input Bit 9 of digital value to DAC 1 1










EXTCH Input Connect ADC channels 0 - 3 to by-pass path around gain stages

1 1

G10 Input Enable 10x gain 0 0

G20 Input Enable 20x gain 0 0

GNDEN Input Ground the negative input to comparator when true

0 0

HOLD Input Sample&Hold signal. Sample analog signal when low. Hold signal when high. If gain stages are used, this signal must go active when ACLK is high.

1 1

IREFEN Input Enables Band-gap reference as AREF signal to DAC

0 0

MUXEN_7 Input Input Mux bit 7 0 0







Note: Incorrect setting of the switches in Figure 24-10 will make signal contention and may damage the part. There are several inputchoices to the S&H circuitry on the negative input of the output comparator in Figure 24-10. Make sure only one path is selectedfrom either one ADC pin, Bandgap reference source, or Ground.



NEGSEL_2 Input Input Mux for negative input for differential signal, bit 2

0 0


0 0


0 0

PASSEN Input Enable pass-gate of gain stages. 1 1

PRECH Input Precharge output latch of comparator. (Active low)

1 1

SCTEST Input Switch-cap TEST enable. Output from x10 gain stage send out to Port Pin having ADC_4

0 0

ST Input Output of gain stages will settle faster if this signal is high first two ACLK periods after AMPEN goes high.

0 0

VCCREN Input Selects Vcc as the ACC reference voltage.

0 0

Table 25-5. Boundary-scan Signals for the ADC (Continued)

Signal Name

Direction as Seenfrom the ADC Description


Output Values when Recommended Inputs are used, and CPU is not Using the ADC


If the ADC is not to be used during scan, the recommended input values from Table 24-5 should be used. The useris recommended not to use the Differential Gain stages during scan. Switch-cap based gain stages require fastoperation and accurate timing which is difficult to obtain when used in a scan chain. Details concerning operationsof the differential gain stage is therefore not provided.The AVR ADC is based on the analog circuitry shown in Figure 24-10 with a successive approximation algorithmimplemented in the digital logic. When used in Boundary-scan, the problem is usually to ensure that an appliedanalog voltage is measured within some limits. This can easily be done without running a successive approxima-tion algorithm: apply the lower limit on the digital DAC[9:0] lines, make sure the output from the comparator is low,then apply the upper limit on the digital DAC[9:0] lines, and verify the output from the comparator to be high.

The ADC need not be used for pure connectivity testing, since all analog inputs are shared with a digital port pin aswell.

When using the ADC, remember the following:

• The Port Pin for the ADC channel in use must be configured to be an input with pull-up disabled to avoid signal contention.

• In Normal mode, a dummy conversion (consisting of 10 comparisons) is performed when enabling the ADC. The user is advised to wait at least 200 ns after enabling the ADC before controlling/observing any ADC signal, or perform a dummy conversion before using the first result.

• The DAC values must be stable at the midpoint value 0x200 when having the HOLD signal low (Sample mode).

As an example, consider the task of verifying a 1.5V ±5% input signal at ADC channel 3 when the power supply is5.0V and AREF is externally connected to VCC.

The recommended values from Table 24-5 are used unless other values are given in the algorithm in Table 24-6.Only the DAC and Port Pin values of the Scan-chain are shown. The column “Actions” describes what JTAGinstruction to be used before filling the Boundary-scan Register with the succeeding columns. The verificationshould be done on the data scanned out when scanning in the data on the same row in the table.

The lower limit is: 1024 1.5V 0,95 5V 291 0x123= = The upper limit is: 1024 1.5V 1.05 5V 323 0x143= =


Using this algorithm, the timing constraint on the HOLD signal constrains the TCK clock frequency. As the algo-rithm keeps HOLD high for five steps, the TCK clock frequency has to be at least five times the number of scan bitsdivided by the maximum hold time, thold,max.

25.6 ATmega32A Boundary-scan OrderTable 24-7 shows the scan order between TDI and TDO when the Boundary-scan chain is selected as data path.Bit 0 is the LSB; the first bit scanned in, and the first bit scanned out. The scan order follows the pin-out order as faras possible. Therefore, the bits of Port A is scanned in the opposite bit order of the other ports. Exceptions from therules are the Scan chains for the analog circuits, which constitute the most significant bits of the scan chain regard-less of which physical pin they are connected to. In Figure 24-3, PXn. Data corresponds to FF0, PXn. Controlcorresponds to FF1, and PXn. Pullup_enable corresponds to FF2. Bit 2, 3, 4, and 5 of Port C is not in the scanchain, since these pins constitute the TAP pins when the JTAG is enabled.

Table 25-6. Algorithm for Using the ADC

Step Actions ADCEN DAC MUXEN HOLD PRECHPA3.Data

PA3.Control

PA3.Pullup_ Enable

1 SAMPLE_PRELOAD

1 0x200 0x08 1 1 0 0 0

2 EXTEST 1 0x200 0x08 0 1 0 0 0

3 1 0x200 0x08 1 1 0 0 0

4 1 0x123 0x08 1 1 0 0 0

5 1 0x123 0x08 1 0 0 0 0

6 Verify the COMP bit scanned out to be 0

1 0x200 0x08 1 1 0 0 0

7 1 0x200 0x08 0 1 0 0 0

8 1 0x200 0x08 1 1 0 0 0

9 1 0x143 0x08 1 1 0 0 0

10 1 0x143 0x08 1 0 0 0 0

11 Verify the COMP bit scanned out to be 1

1 0x200 0x08 1 1 0 0 0

Table 25-7. ATmega32A Boundary-scan Order

Bit Number Signal Name Module

140 AC_IDLE Comparator

139 ACO

138 ACME

137 ACBG


136 COMP ADC

135 PRIVATE_SIGNAL1(1)

134 ACLK

133 ACTEN

132 PRIVATE_SIGNAL2(2)

131 ADCBGEN

130 ADCEN

129 AMPEN

128 DAC_9

127 DAC_8

126 DAC_7

125 DAC_6

124 DAC_5

123 DAC_4

122 DAC_3

121 DAC_2

120 DAC_1

119 DAC_0

118 EXTCH

117 G10

116 G20

115 GNDEN

114 HOLD

113 IREFEN

Table 25-7. ATmega32A Boundary-scan Order (Continued)



112 MUXEN_7 ADC

111 MUXEN_6

110 MUXEN_5

109 MUXEN_4

108 MUXEN_3

107 MUXEN_2

106 MUXEN_1

105 MUXEN_0

104 NEGSEL_2

103 NEGSEL_1

102 NEGSEL_0

101 PASSEN

100 PRECH

99 SCTEST

98 ST

97 VCCREN

96 PB0.Data Port B

95 PB0.Control

94 PB0.Pullup_Enable

93 PB1.Data

92 PB1.Control


90 PB2.Data

89 PB2.Control


87 PB3.Data

86 PB3.Control


84 PB4.Data

83 PB4.Control





81 PB5.Data Port B

80 PB5.Control


78 PB6.Data

77 PB6.Control


75 PB7.Data

74 PB7.Control


72 RSTT Reset Logic

(Observe-Only)71 RSTHV

70 EXTCLKEN Enable signals for main clock/Oscillators

69 OSCON

68 RCOSCEN

67 OSC32EN

66 EXTCLK (XTAL1) Clock input and Oscillators for the main clock

(Observe-Only)65 OSCCK

64 RCCK

63 OSC32CK

62 TWIEN TWI

61 PD0.Data Port D

60 PD0.Control

59 PD0.Pullup_Enable

58 PD1.Data

57 PD1.Control


55 PD2.Data

54 PD2.Control


52 PD3.Data

51 PD3.Control


49 PD4.Data

48 PD4.Control





46 PD5.Data Port D

45 PD5.Control


43 PD6.Data

42 PD6.Control


40 PD7.Data

39 PD7.Control


37 PC0.Data Port C

36 PC0.Control

35 PC0.Pullup_Enable

34 PC1.Data

33 PC1.Control


31 PC6.Data

30 PC6.Control


28 PC7.Data

27 PC7.Control


25 TOSC 32kHz Timer Oscillator

24 TOSCON

23 PA7.Data Port A

22 PA7.Control

21 PA7.Pullup_Enable

20 PA6.Data

19 PA6.Control


17 PA5.Data

16 PA5.Control


14 PA4.Data

13 PA4.Control





Notes: 1. PRIVATE_SIGNAL1 should always be scanned in as zero.2. PRIVATE_SIGNAL2 should always be scanned in as zero.

25.7 Boundary-scan Description Language FilesBoundary-scan Description Language (BSDL) files describe Boundary-scan capable devices in a standard formatused by automated test-generation software. The order and function of bits in the Boundary-scan Data Registerare included in this description. A BSDL file for ATmega32A is available.


25.8.1 MCU Control and Status Register – MCUCSRThe MCU Control and Status Register contains control bits for general MCU functions, and provides information onwhich reset source caused an MCU Reset.

• Bit 7 – JTD: JTAG Interface DisableWhen this bit is zero, the JTAG interface is enabled if the JTAGEN Fuse is programmed. If this bit is one, the JTAGinterface is disabled. In order to avoid unintentional disabling or enabling of the JTAG interface, a timed sequencemust be followed when changing this bit: The application software must write this bit to the desired value twicewithin four cycles to change its value.

If the JTAG interface is left unconnected to other JTAG circuitry, the JTD bit should be set to one. The reason forthis is to avoid static current at the TDO pin in the JTAG interface.

• Bit 4 – JTRF: JTAG Reset FlagThis bit is set if a reset is being caused by a logic one in the JTAG Reset Register selected by the JTAG instructionAVR_RESET. This bit is reset by a Power-on Reset, or by writing a logic zero to the flag.

11 PA3.Data Port A

10 PA3.Control

9 PA3.Pullup_Enable

8 PA2.Data

7 PA2.Control

6 PA2.Pullup_Enable

5 PA1.Data

4 PA1.Control

3 PA1.Pullup_Enable

2 PA0.Data

1 PA0.Control

0 PA0.Pullup_Enable



Bit 7 6 5 4 3 2 1 0





26. Boot Loader Support – Read-While-Write Self-Programming

26.1 Features• Read-While-Write Self-Programming• Flexible Boot Memory size• High Security (Separate Boot Lock Bits for a Flexible Protection)• Separate Fuse to Select Reset Vector• Optimized Page(1) Size• Code Efficient Algorithm• Efficient Read-Modify-Write Support

Note: 1. A page is a section in the flash consisting of several bytes (see Table 26-5 on page 269) used during program-ming. The page organization does not affect normal operation.

26.2 OverviewThe Boot Loader Support provides a real Read-While-Write Self-Programming mechanism for downloading anduploading program code by the MCU itself. This feature allows flexible application software updates controlled bythe MCU using a Flash-resident Boot Loader program. The Boot Loader program can use any available data inter-face and associated protocol to read code and write (program) that code into the Flash memory, or read the codefrom the Program memory. The program code within the Boot Loader section has the capability to write into theentire Flash, including the Boot Loader memory. The Boot Loader can thus even modify itself, and it can also eraseitself from the code if the feature is not needed anymore. The size of the Boot Loader memory is configurable withFuses and the Boot Loader has two separate sets of Boot Lock bits which can be set independently. This gives theuser a unique flexibility to select different levels of protection.

26.3 Application and Boot Loader Flash SectionsThe Flash memory is organized in two main sections, the Application section and the Boot Loader section (see Fig-ure 25-2). The size of the different sections is configured by the BOOTSZ Fuses as shown in Table 25-6 on page263 and Figure 25-2. These two sections can have different level of protection since they have different sets ofLock bits.

26.3.1 Application SectionThe Application section is the section of the Flash that is used for storing the application code. The protection levelfor the application section can be selected by the Application Boot Lock bits (Boot Lock bits 0), see Table 25-2 onpage 256. The Application section can never store any Boot Loader code since the SPM instruction is disabledwhen executed from the Application section.

26.3.2 BLS – Boot Loader SectionWhile the Application section is used for storing the application code, the The Boot Loader software must belocated in the BLS since the SPM instruction can initiate a programming when executing from the BLS only. TheSPM instruction can access the entire Flash, including the BLS itself. The protection level for the Boot Loader sec-tion can be selected by the Boot Loader Lock bits (Boot Lock bits 1), see Table 25-3 on page 256.

26.4 Read-While-Write and no Read-While-Write Flash SectionsWhether the CPU supports Read-While-Write or if the CPU is halted during a Boot Loader software update isdependent on which address that is being programmed. In addition to the two sections that are configurable by theBOOTSZ Fuses as described above, the Flash is also divided into two fixed sections, the Read-While-Write(RWW) section and the No Read-While-Write (NRWW) section. The limit between the RWW- and NRWW sectionsis given in Table 25-7 on page 263 and Figure 25-2 on page 255. The main difference between the two sections is:


• When erasing or writing a page located inside the RWW section, the NRWW section can be read during the operation.

• When erasing or writing a page located inside the NRWW section, the CPU is halted during the entire operation.

Note that the user software can never read any code that is located inside the RWW section during a Boot Loadersoftware operation. The syntax “Read-While-Write section” refers to which section that is being programmed(erased or written), not which section that actually is being read during a Boot Loader software update.

26.4.1 RWW – Read-While-Write SectionIf a Boot Loader software update is programming a page inside the RWW section, it is possible to read code fromthe Flash, but only code that is located in the NRWW section. During an on-going programming, the software mustensure that the RWW section never is being read. If the user software is trying to read code that is located insidethe RWW section (that is, by a call/jmp/lpm or an interrupt) during programming, the software might end up in anunknown state. To avoid this, the interrupts should either be disabled or moved to the Boot Loader section. TheBoot Loader section is always located in the NRWW section. The RWW Section Busy bit (RWWSB) in the StoreProgram Memory Control Register (SPMCR) will be read as logical one as long as the RWW section is blocked forreading. After a programming is completed, the RWWSB must be cleared by software before reading code locatedin the RWW section. See “SPMCR – Store Program Memory Control Register” on page 264. for details on how toclear RWWSB.

26.4.2 NRWW – No Read-While-Write SectionThe code located in the NRWW section can be read when the Boot Loader software is updating a page in theRWW section. When the Boot Loader code updates the NRWW section, the CPU is halted during the entire pageerase or page write operation.

Table 26-1. Read-While-Write Features

Which Section does the Z-pointer Address during the

Programming?

Which Section can be Read during

Programming?Is the CPU

Halted?

Read-While-Write

Supported?

RWW section NRWW section No Yes

NRWW section None Yes No


Figure 26-1. Read-While-Write vs. No Read-While-Write

Read-While-Write(RWW) Section

No Read-While-Write (NRWW) Section

Z-pointerAddresses RWWSection

Z-pointerAddresses NRWWSection

CPU is Haltedduring the Operation

Code Located in NRWW SectionCan be Read duringthe Operation


Figure 26-2. Memory Sections(1)

Note: 1. The parameters in the figure above are given in Table 25-6 on page 263.

26.5 Boot Loader Lock BitsIf no Boot Loader capability is needed, the entire Flash is available for application code. The Boot Loader has twoseparate sets of Boot Lock bits which can be set independently. This gives the user a unique flexibility to select dif-ferent levels of protection.

The user can select:

• To protect the entire Flash from a software update by the MCU

• To protect only the Boot Loader Flash section from a software update by the MCU

• To protect only the Application Flash section from a software update by the MCU

• Allow software update in the entire Flash

See Table 25-2 and Table 25-3 for further details. The Boot Lock bits can be set in software and in Serial or Paral-lel Programming mode, but they can be cleared by a Chip Erase command only. The general Write Lock (Lock Bitmode 2) does not control the programming of the Flash memory by SPM instruction. Similarly, the generalRead/Write Lock (Lock Bit mode 3) does not control reading nor writing by LPM/SPM, if it is attempted.

$0000

Flashend

Program MemoryBOOTSZ = '11'


Boot Loader Flash SectionFlashend


$0000




Boot Loader Flash Section

$0000

Flashend


Flashend

End RWW

Start NRWW

Application flash Section



End RWW

Start NRWW

End RWW

Start NRWW

$0000

End RWW, End Application

Start NRWW, Start Boot Loader

Application Flash SectionApplication Flash Section


Rea

d-W

hile

-Writ

e S

ectio

nN

o R

ead-

Whi

le-W

rite

Sec

tion

Rea

d-W

hile

-Writ

e S

ectio

nN

o R

ead-

Whi

le-W

rite

Sec

tion

Rea

d-W

hile

-Writ

e S

ectio

nN

o R

ead-

Whi

le-W

rite

Sec

tion

Rea

d-W

hile

-Writ

e S

ectio

nN

o R

ead-

Whi

le-W

rite

Sec

tion

End Application

Start Boot Loader

End Application

Start Boot Loader

End Application

Start Boot Loader


Note: 1. “1” means unprogrammed, “0” means programmed


26.6 Entering the Boot Loader ProgramEntering the Boot Loader takes place by a jump or call from the application program. This may be initiated by a trig-ger such as a command received via USART, or SPI interface. Alternatively, the Boot Reset Fuse can beprogrammed so that the Reset Vector is pointing to the Boot Flash start address after a reset. In this case, the BootLoader is started after a reset. After the application code is loaded, the program can start executing the applicationcode. Note that the fuses cannot be changed by the MCU itself. This means that once the Boot Reset Fuse is pro-grammed, the Reset Vector will always point to the Boot Loader Reset and the fuse can only be changed throughthe serial or parallel programming interface.

Table 26-2. Boot Lock Bit0 Protection Modes (Application Section)(1)

BLB0 Mode BLB02 BLB01 Protection

1 1 1No restrictions for SPM or LPM accessing the Application section.

2 1 0 SPM is not allowed to write to the Application section.

3 0 0

SPM is not allowed to write to the Application section, and LPM executing from the Boot Loader section is not allowed to read from the Application section. If interrupt vectors are placed in the Boot Loader section, interrupts are disabled while executing from the Application section.

4 0 1

LPM executing from the Boot Loader section is not allowed to read from the Application section. If interrupt vectors are placed in the Boot Loader section, interrupts are disabled while executing from the Application section.

Table 26-3. Boot Lock Bit1 Protection Modes (Boot Loader Section)(1)

BLB1 mode BLB12 BLB11 Protection

1 1 1No restrictions for SPM or LPM accessing the Boot Loader section.

2 1 0 SPM is not allowed to write to the Boot Loader section.

3 0 0

SPM is not allowed to write to the Boot Loader section, and LPM executing from the Application section is not allowed to read from the Boot Loader section. If interrupt vectors are placed in the Application section, interrupts are disabled while executing from the Boot Loader section.

4 0 1

LPM executing from the Application section is not allowed to read from the Boot Loader section. If interrupt vectors are placed in the Application section, interrupts are disabled while executing from the Boot Loader section.



26.7 Addressing the Flash during Self-ProgrammingThe Z-pointer is used to address the SPM commands.

Since the Flash is organized in pages (see Table 26-5 on page 269), the Program Counter can be treated as hav-ing two different sections. One section, consisting of the least significant bits, is addressing the words within apage, while the most significant bits are addressing the pages. This is shown in Figure 25-3. Note that the PageErase and Page Write operations are addressed independently. Therefore it is of major importance that the BootLoader software addresses the same page in both the Page Erase and Page Write operation. Once a program-ming operation is initiated, the address is latched and the Z-pointer can be used for other operations.

The only SPM operation that does not use the Z-pointer is Setting the Boot Loader Lock bits. The content of the Z-pointer is ignored and will have no effect on the operation. The LPM instruction does also use the Z pointer to storethe address. Since this instruction addresses the Flash byte by byte, also the LSB (bit Z0) of the Z-pointer is used.

Figure 26-3. Addressing the Flash during SPM(1)

Notes: 1. The different variables used in Figure 25-3 are listed in Table 25-8 on page 263.

2. PCPAGE and PCWORD are listed in “Page Size” on page 269.

Table 26-4. Boot Reset Fuse(1)

BOOTRST Reset Address

1 Reset Vector = Application reset (address $0000)

0 Reset Vector = Boot Loader reset (see Table 25-6 on page 263)

Bit 15 14 13 12 11 10 9 8

ZH (R31) Z15 Z14 Z13 Z12 Z11 Z10 Z9 Z8

ZL (R30) Z7 Z6 Z5 Z4 Z3 Z2 Z1 Z0

7 6 5 4 3 2 1 0

PROGRAM MEMORY

0115

Z - REGISTER

BIT

0

ZPAGEMSB

WORD ADDRESSWITHIN A PAGE

PAGE ADDRESSWITHIN THE FLASH

ZPCMSB

INSTRUCTION WORD

PAGE PCWORD[PAGEMSB:0]:

00

01

02

PAGEEND

PAGE

PCWORDPCPAGE

PCMSB PAGEMSBPROGRAMCOUNTER


26.8 Self-Programming the FlashThe program memory is updated in a page by page fashion. Before programming a page with the data stored in thetemporary page buffer, the page must be erased. The temporary page buffer is filled one word at a time using SPMand the buffer can be filled either before the page erase command or between a page erase and a page writeoperation:

Alternative 1, fill the buffer before a Page Erase

• Fill temporary page buffer

• Perform a Page Erase

• Perform a Page Write

Alternative 2, fill the buffer after Page Erase

• Perform a Page Erase

• Fill temporary page buffer

• Perform a Page Write

If only a part of the page needs to be changed, the rest of the page must be stored (for example in the temporarypage buffer) before the erase, and then be rewritten. When using alternative 1, the Boot Loader provides an effec-tive Read-Modify-Write feature which allows the user software to first read the page, do the necessary changes,and then write back the modified data. If alternative 2 is used, it is not possible to read the old data while loadingsince the page is already erased. The temporary page buffer can be accessed in a random sequence. It is essen-tial that the page address used in both the page erase and page write operation is addressing the same page. See“Simple Assembly Code Example for a Boot Loader” on page 261 for an assembly code example.

26.8.1 Performing Page Erase by SPMTo execute Page Erase, set up the address in the Z-pointer, write “X0000011” to SPMCR and execute SPM withinfour clock cycles after writing SPMCR. The data in R1 and R0 is ignored. The page address must be written toPCPAGE in the Z-register. Other bits in the Z-pointer must be written zero during this operation.

• Page Erase to the RWW section: The NRWW section can be read during the page erase.

• Page Erase to the NRWW section: The CPU is halted during the operation.

Note: If an iterrupt occurs in the timed sequence, the four cycle access cannot be guaranteed. In order to ensure atomicoperation disable interrupts before writing to SPMCSR.

26.8.2 Filling the Temporary Buffer (Page Loading)To write an instruction word, set up the address in the Z-pointer and data in R1:R0, write “00000001” to SPMCRand execute SPM within four clock cycles after writing SPMCR. The content of PCWORD in the Z-register is usedto address the data in the temporary buffer. The temporary buffer will auto-erase after a page write operation or bywriting the RWWSRE bit in SPMCR. It is also erased after a system reset. Note that it is not possible to write morethan one time to each address without erasing the temporary buffer.

Note: If the EEPROM is written in the middle of an SPM Page Load operation, all data loaded will be lost.

26.8.3 Performing a Page WriteTo execute Page Write, set up the address in the Z-pointer, write “X0000101” to SPMCR and execute SPM withinfour clock cycles after writing SPMCR. The data in R1 and R0 is ignored. The page address must be written toPCPAGE. Other bits in the Z-pointer must be written to zero during this operation.

• Page Write to the RWW section: The NRWW section can be read during the Page Write.

• Page Write to the NRWW section: The CPU is halted during the operation.


26.8.4 Using the SPM InterruptIf the SPM interrupt is enabled, the SPM interrupt will generate a constant interrupt when the SPMEN bit inSPMCR is cleared. This means that the interrupt can be used instead of polling the SPMCR Register in software.When using the SPM interrupt, the Interrupt Vectors should be moved to the BLS section to avoid that an interruptis accessing the RWW section when it is blocked for reading. How to move the interrupts is described in “Inter-rupts” on page 45.

26.8.5 Consideration while Updating BLSSpecial care must be taken if the user allows the Boot Loader section to be updated by leaving Boot Lock bit11unprogrammed. An accidental write to the Boot Loader itself can corrupt the entire Boot Loader, and further soft-ware updates might be impossible. If it is not necessary to change the Boot Loader software itself, it isrecommended to program the Boot Lock bit11 to protect the Boot Loader software from any internal softwarechanges.

26.8.6 Prevent Reading the RWW Section during Self-ProgrammingDuring Self-Programming (either Page Erase or Page Write), the RWW section is always blocked for reading. Theuser software itself must prevent that this section is addressed during the Self-Programming operation. TheRWWSB in the SPMCR will be set as long as the RWW section is busy. During self-programming the InterruptVector table should be moved to the BLS as described in “Interrupts” on page 45, or the interrupts must be dis-abled. Before addressing the RWW section after the programming is completed, the user software must clear theRWWSB by writing the RWWSRE. See “Simple Assembly Code Example for a Boot Loader” on page 261 for anexample.

26.8.7 Setting the Boot Loader Lock Bits by SPMTo set the Boot Loader Lock bits, write the desired data to R0, write “X0001001” to SPMCR and execute SPMwithin four clock cycles after writing SPMCR. The only accessible Lock bits are the Boot Lock bits that may preventthe Application and Boot Loader section from any software update by the MCU.

See Table 25-2 and Table 25-3 for how the different settings of the Boot Loader bits affect the Flash access.

If bits 5:2 in R0 are cleared (zero), the corresponding Boot Lock bit will be programmed if an SPM instruction isexecuted within four cycles after BLBSET and SPMEN are set in SPMCR. The Z-pointer is don’t care during thisoperation, but for future compatibility it is recommended to load the Z-pointer with $0001 (same as used for readingthe Lock bits). For future compatibility It is also recommended to set bits 7, 6, 1, and 0 in R0 to “1” when writing theLock bits. When programming the Lock bits the entire Flash can be read during the operation.

26.8.8 EEPROM Write Prevents Writing to SPMCRNote that an EEPROM write operation will block all software programming to Flash. Reading the Fuses and Lockbits from software will also be prevented during the EEPROM write operation. It is recommended that the userchecks the status bit (EEWE) in the EECR Register and verifies that the bit is cleared before writing to the SPMCRRegister.

26.8.9 Reading the Fuse and Lock Bits from SoftwareIt is possible to read both the Fuse and Lock bits from software. To read the Lock bits, load the Z-pointer with$0001 and set the BLBSET and SPMEN bits in SPMCR. When an LPM instruction is executed within three CPUcycles after the BLBSET and SPMEN bits are set in SPMCR, the value of the Lock bits will be loaded in the desti-nation register. The BLBSET and SPMEN bits will auto-clear upon completion of reading the Lock bits or if no LPM

Bit 7 6 5 4 3 2 1 0

R0 1 1 BLB12 BLB11 BLB02 BLB01 1 1


instruction is executed within three CPU cycles or no SPM instruction is executed within four CPU cycles. WhenBLBSET and SPMEN are cleared, LPM will work as described in the Instruction set Manual.

The algorithm for reading the Fuse Low bits is similar to the one described above for reading the Lock bits. To readthe Fuse Low bits, load the Z-pointer with $0000 and set the BLBSET and SPMEN bits in SPMCR. When an LPMinstruction is executed within three cycles after the BLBSET and SPMEN bits are set in the SPMCR, the value ofthe Fuse Low bits (FLB) will be loaded in the destination register as shown below. Refer to Table 26-4 on page 268for a detailed description and mapping of the Fuse Low bits.

Similarly, when reading the Fuse High bits, load $0003 in the Z-pointer. When an LPM instruction is executedwithin three cycles after the BLBSET and SPMEN bits are set in the SPMCR, the value of the Fuse High bits (FHB)will be loaded in the destination register as shown below. Refer to Table 26-3 on page 267 for detailed descriptionand mapping of the Fuse High bits.

Fuse and Lock bits that are programmed, will be read as zero. Fuse and Lock bits that are unprogrammed, will beread as one.

26.8.10 Preventing Flash CorruptionDuring periods of low VCC, the Flash program can be corrupted because the supply voltage is too low for the CPUand the Flash to operate properly. These issues are the same as for board level systems using the Flash, and thesame design solutions should be applied.

A Flash program corruption can be caused by two situations when the voltage is too low. First, a regular writesequence to the Flash requires a minimum voltage to operate correctly. Secondly, the CPU itself can executeinstructions incorrectly, if the supply voltage for executing instructions is too low.

Flash corruption can easily be avoided by following these design recommendations (one is sufficient):

1. If there is no need for a Boot Loader update in the system, program the Boot Loader Lock bits to prevent any Boot Loader software updates.

2. Keep the AVR RESET active (low) during periods of insufficient power supply voltage. This can be done by enabling the internal Brown-out Detector (BOD) if the operating voltage matches the detection level. If not, an external low VCC Reset Protection circuit can be used. If a reset occurs while a write operation is in progress, the write operation will be completed provided that the power supply voltage is sufficient.

3. Keep the AVR core in Power-down Sleep mode during periods of low VCC. This will prevent the CPU from attempting to decode and execute instructions, effectively protecting the SPMCR Register and thus the Flash from unintentional writes.

26.8.11 Programming Time for Flash when using SPMThe Calibrated RC Oscillator is used to time Flash accesses. Table 25-5 shows the typical programming time forFlash accesses from the CPU.

Bit 7 6 5 4 3 2 1 0

Rd – – BLB12 BLB11 BLB02 BLB01 LB2 LB1

Bit 7 6 5 4 3 2 1 0

Rd FLB7 FLB6 FLB5 FLB4 FLB3 FLB2 FLB1 FLB0

Bit 7 6 5 4 3 2 1 0

Rd FHB7 FHB6 FHB5 FHB4 FHB3 FHB2 FHB1 FHB0

Table 26-5. SPM Programming Time.

Symbol Min Programming Time Max Programming Time

Flash write (Page Erase, Page Write, and write Lock bits by SPM)

3.7ms 4.5ms


26.8.12 Simple Assembly Code Example for a Boot Loader;-the routine writes one page of data from RAM to Flash; the first data location in RAM is pointed to by the Y pointer; the first data location in Flash is pointed to by the Z pointer;-error handling is not included;-the routine must be placed inside the boot space; (at least the Do_spm sub routine). Only code inside NRWW section can; be read during self-programming (page erase and page write).;-registers used: r0, r1, temp1 (r16), temp2 (r17), looplo (r24), ; loophi (r25), spmcrval (r20); storing and restoring of registers is not included in the routine; register usage can be optimized at the expense of code size;-It is assumed that either the interrupt table is moved to the Boot; loader section or that the interrupts are disabled.

.equ PAGESIZEB = PAGESIZE*2 ; PAGESIZEB is page size in BYTES, not; words

.org SMALLBOOTSTARTWrite_page:; page eraseldi spmcrval, (1<<PGERS) | (1<<SPMEN)call Do_spm

; re-enable the RWW sectionldi spmcrval, (1<<RWWSRE) | (1<<SPMEN)call Do_spm

; transfer data from RAM to Flash page bufferldi looplo, low(PAGESIZEB) ;init loop variableldi loophi, high(PAGESIZEB) ;not required for PAGESIZEB<=256

Wrloop:ld r0, Y+ld r1, Y+ldi spmcrval, (1<<SPMEN)call Do_spmadiw ZH:ZL, 2sbiw loophi:looplo, 2 ;use subi for PAGESIZEB<=256brne Wrloop

; execute page writesubi ZL, low(PAGESIZEB) ;restore pointersbci ZH, high(PAGESIZEB) ;not required for PAGESIZEB<=256ldi spmcrval, (1<<PGWRT) | (1<<SPMEN)call Do_spm

; re-enable the RWW sectionldi spmcrval, (1<<RWWSRE) | (1<<SPMEN)call Do_spm

; read back and check, optionalldi looplo, low(PAGESIZEB) ;init loop variableldi loophi, high(PAGESIZEB) ;not required for PAGESIZEB<=256subi YL, low(PAGESIZEB) ;restore pointersbci YH, high(PAGESIZEB)

Rdloop:lpm r0, Z+ld r1, Y+cpse r0, r1jmp Errorsbiw loophi:looplo, 1 ;use subi for PAGESIZEB<=256brne Rdloop


; return to RWW section; verify that RWW section is safe to read

Return:in temp1, SPMCRsbrs temp1, RWWSB ; If RWWSB is set, the RWW section is not

; ready yetret; re-enable the RWW sectionldi spmcrval, (1<<RWWSRE) | (1<<SPMEN)call Do_spmrjmp Return

Do_spm:; check for previous SPM complete

Wait_spm:in temp1, SPMCRsbrc temp1, SPMENrjmp Wait_spm; input: spmcrval determines SPM action; disable interrupts if enabled, store statusin temp2, SREGcli; check that no EEPROM write access is present

Wait_ee:sbic EECR, EEWErjmp Wait_ee; SPM timed sequenceout SPMCR, spmcrvalspm; restore SREG (to enable interrupts if originally enabled)out SREG, temp2ret

26.8.13 Boot Loader ParametersIn Table 25-6 through Table 25-8, the parameters used in the description of the self programming are given.

Note: 1. The different BOOTSZ Fuse configurations are shown in Figure 25-2

Table 26-6. Boot Size Configuration(1)

BOOTSZ1 BOOTSZ0Boot Size Pages



End Application section

Boot Reset Address (start Boot Loader Section)

1 1256 words

4$0000 - $3EFF

$3F00 - $3FFF

$3EFF $3F00

1 0512 words

8$0000 - $3DFF

$3E00 - $3FFF

$3DFF $3E00

0 11024 words

16$0000 - $3BFF

$3C00 - $3FFF

$3BFF $3C00

0 02048 words

32$0000 - $37FF

$3800 - $3FFF

$37FF $3800

Table 26-7. Read-While-Write Limit(1)

Section Pages Address

Read-While-Write section (RWW) 224 $0000 - $37FF

No Read-While-Write section (NRWW) 32 $3800 - $3FFF


Note: 1. For details about these two section, see “NRWW – No Read-While-Write Section” on page 253 and “RWW – Read-While-Write Section” on page 253

Note: 1. Z15: always ignoredZ0: should be zero for all SPM commands, byte select for the LPM instruction.See “Addressing the Flash during Self-Programming” on page 257 for details about the use of Z-pointer during Self-Programming.


26.9.1 SPMCR – Store Program Memory Control RegisterThe Store Program Memory Control Register contains the control bits needed to control the Boot Loaderoperations.

• Bit 7 – SPMIE: SPM Interrupt EnableWhen the SPMIE bit is written to one, and the I-bit in the Status Register is set (one), the SPM ready interrupt willbe enabled. The SPM ready Interrupt will be executed as long as the SPMEN bit in the SPMCR Register is cleared.

• Bit 6 – RWWSB: Read-While-Write Section BusyWhen a self-programming (Page Erase or Page Write) operation to the RWW section is initiated, the RWWSB willbe set (one) by hardware. When the RWWSB bit is set, the RWW section cannot be accessed. The RWWSB bitwill be cleared if the RWWSRE bit is written to one after a Self-Programming operation is completed. Alternativelythe RWWSB bit will automatically be cleared if a page load operation is initiated.

• Bit 5 – Reserved BitThis bit is a reserved bit in the ATmega32A and always read as zero.

• Bit 4 – RWWSRE: Read-While-Write Section Read Enable

Table 26-8. Explanation of Different Variables used in Figure 25-3 and the Mapping to the Z-pointer

VariableCorresponding

Z-value(1) Description

PCMSB13 Most significant bit in the Program Counter. (The

Program Counter is 14 bits PC[13:0])

PAGEMSB5 Most significant bit which is used to address the

words within one page (64 words in a page requires 6 bits PC [5:0]).

ZPCMSBZ14 Bit in Z-register that is mapped to PCMSB. Because

Z0 is not used, the ZPCMSB equals PCMSB + 1.

ZPAGEMSBZ6 Bit in Z-register that is mapped to PAGEMSB.

Because Z0 is not used, the ZPAGEMSB equals PAGEMSB + 1.

PCPAGEPC[13:6] Z14:Z7 Program Counter page address: Page select, for

page erase and page write

PCWORDPC[5:0] Z6:Z1 Program Counter word address: Word select, for

filling temporary buffer (must be zero during page write operation)

Bit 7 6 5 4 3 2 1 0

SPMIE RWWSB – RWWSRE BLBSET PGWRT PGERS SPMEN SPMCR

Read/Write R/W R R R/W R/W R/W R/W R/W

Initial value 0 0 0 0 0 0 0 0


When programming (Page Erase or Page Write) to the RWW section, the RWW section is blocked for reading (theRWWSB will be set by hardware). To re-enable the RWW section, the user software must wait until the program-ming is completed (SPMEN will be cleared). Then, if the RWWSRE bit is written to one at the same time asSPMEN, the next SPM instruction within four clock cycles re-enables the RWW section. The RWW section cannotbe re-enabled while the Flash is busy with a page erase or a page write (SPMEN is set). If the RWWSRE bit is writ-ten while the Flash is being loaded, the Flash load operation will abort and the data loaded will be lost.

• Bit 3 – BLBSET: Boot Lock Bit SetIf this bit is written to one at the same time as SPMEN, the next SPM instruction within four clock cycles sets BootLock bits, according to the data in R0. The data in R1 and the address in the Z-pointer are ignored. The BLBSETbit will automatically be cleared upon completion of the Lock bit set, or if no SPM instruction is executed within fourclock cycles.

An LPM instruction within three cycles after BLBSET and SPMEN are set in the SPMCR Register, will read eitherthe Lock bits or the Fuse bits (depending on Z0 in the Z-pointer) into the destination register. See “Reading theFuse and Lock Bits from Software” on page 260 for details.

• Bit 2 – PGWRT: Page WriteIf this bit is written to one at the same time as SPMEN, the next SPM instruction within four clock cycles executesPage Write, with the data stored in the temporary buffer. The page address is taken from the high part of the Z-pointer. The data in R1 and R0 are ignored. The PGWRT bit will auto-clear upon completion of a page write, or ifno SPM instruction is executed within four clock cycles. The CPU is halted during the entire page write operation ifthe NRWW section is addressed.

• Bit 1 – PGERS: Page EraseIf this bit is written to one at the same time as SPMEN, the next SPM instruction within four clock cycles executesPage Erase. The page address is taken from the high part of the Z-pointer. The data in R1 and R0 are ignored. ThePGERS bit will auto-clear upon completion of a page erase, or if no SPM instruction is executed within four clockcycles. The CPU is halted during the entire page write operation if the NRWW section is addressed.

• Bit 0 – SPMEN: Store Program Memory EnableThis bit enables the SPM instruction for the next four clock cycles. If written to one together with either RWWSRE,BLBSET, PGWRT’ or PGERS, the following SPM instruction will have a special meaning, see description above. Ifonly SPMEN is written, the following SPM instruction will store the value in R1:R0 in the temporary page bufferaddressed by the Z-pointer. The LSB of the Z-pointer is ignored. The SPMEN bit will auto-clear upon completion ofan SPM instruction, or if no SPM instruction is executed within four clock cycles. During page erase and pagewrite, the SPMEN bit remains high until the operation is completed.

Writing any other combination than “10001”, “01001”, “00101”, “00011” or “00001” in the lower five bits will have noeffect.


27. Memory Programming

27.1 Program And Data Memory Lock BitsThe ATmega32A provides six Lock bits which can be left unprogrammed (“1”) or can be programmed (“0”) toobtain the additional features listed in Table 26-2. The Lock bits can only be erased to “1” with the Chip Erasecommand.


Table 27-1. Lock Bit Byte(1)

Lock Bit Byte Bit No. Description Default Value

7 – 1 (unprogrammed)

6 – 1 (unprogrammed)

BLB12 5 Boot Lock bit 1 (unprogrammed)




LB2 1 Lock bit 1 (unprogrammed)

LB1 0 Lock bit 1 (unprogrammed)

Table 27-2. Lock Bit Protection Modes

Memory Lock Bits(2) Protection Type

LB Mode LB2 LB1

1 1 1 No memory lock features enabled.

2 1 0Further programming of the Flash and EEPROM is disabled in Parallel and SPI/JTAG Serial Programming mode. The Fuse bits are locked in both Serial and Parallel Programming mode.(1)

3 0 0

Further programming and verification of the Flash and EEPROM is disabled in Parallel and SPI/JTAG Serial Programming mode. The Fuse bits are locked in both Serial and Parallel Programming mode.(1)

BLB0 Mode BLB02 BLB01

1 1 1No restrictions for SPM or LPM accessing the Application section.

2 1 0 SPM is not allowed to write to the Application section.

3 0 0

SPM is not allowed to write to the Application section, and LPM executing from the Boot Loader section is not allowed to read from the Application section. If interrupt vectors are placed in the Boot Loader section, interrupts are disabled while executing from the Application section.

4 0 1

LPM executing from the Boot Loader section is not allowed to read from the Application section. If interrupt vectors are placed in the Boot Loader section, interrupts are disabled while executing from the Application section.

BLB1 Mode BLB12 BLB11


Notes: 1. Program the fuse bits before programming the Lock bits.

2. “1” means unprogrammed, “0” means programmed

27.2 Fuse BitsThe ATmega32A has two fuse bytes. Table 26-3 and Table 26-4 describe briefly the functionality of all the fusesand how they are mapped into the fuse bytes. Note that the fuses are read as logical zero, “0”, if they areprogrammed.

Notes: 1. The SPIEN Fuse is not accessible in SPI Serial Programming mode.

2. The CKOPT Fuse functionality depends on the setting of the CKSEL bits. See See “Clock Sources” on page 26. for details.

3. The default value of BOOTSZ1:0 results in maximum Boot Size. See Table 25-6 on page 263.

4. Never ship a product with the OCDEN Fuse programmed regardless of the setting of Lock bits and the JTAGEN Fuse. A programmed OCDEN Fuse enables some parts of the clock system to be running in all sleep modes. This may increase the power consumption.

5. If the JTAG interface is left unconnected, the JTAGEN fuse should if possible be disabled. This to avoid static cur-rent at the TDO pin in the JTAG interface.

1 1 1No restrictions for SPM or LPM accessing the Boot Loader section.

2 1 0 SPM is not allowed to write to the Boot Loader section.

3 0 0

SPM is not allowed to write to the Boot Loader section, and LPM executing from the Application section is not allowed to read from the Boot Loader section. If interrupt vectors are placed in the Application section, interrupts are disabled while executing from the Boot Loader section.

4 0 1

LPM executing from the Application section is not allowed to read from the Boot Loader section. If interrupt vectors are placed in the Application section, interrupts are disabled while executing from the Boot Loader section.

Table 27-2. Lock Bit Protection Modes (Continued)

Memory Lock Bits(2) Protection Type

Table 27-3. Fuse High Byte

Fuse High Byte Bit No. Description Default Value

OCDEN(4) 7 Enable OCD 1 (unprogrammed, OCD disabled)

JTAGEN(5) 6 Enable JTAG 0 (programmed, JTAG enabled)

SPIEN(1) 5Enable SPI Serial Program and Data Downloading

0 (programmed, SPI prog. enabled)

CKOPT(2) 4 Oscillator options 1 (unprogrammed)

EESAVE 3EEPROM memory is preserved through the Chip Erase

1 (unprogrammed, EEPROM not preserved)

BOOTSZ1 2Select Boot Size (see Table 25-6 for details) 0 (programmed)(3)

BOOTSZ0 1Select Boot Size (see Table 25-6 for details) 0 (programmed)(3)

BOOTRST 0 Select reset vector 1 (unprogrammed)


Notes: 1. The default value of SUT1:0 results in maximum start-up time. See Table 8-9 on page 30 for details.

2. The default setting of CKSEL3:0 results in internal RC Oscillator @ 1MHz. See Table 8-1 on page 26 for details.

The status of the Fuse bits is not affected by Chip Erase. Note that the Fuse bits are locked if Lock bit1 (LB1) isprogrammed. Program the Fuse bits before programming the Lock bits.

27.2.1 Latching of FusesThe Fuse values are latched when the device enters programming mode and changes of the Fuse values will haveno effect until the part leaves Programming mode. This does not apply to the EESAVE Fuse which will take effectonce it is programmed. The fuses are also latched on Power-up in Normal mode.

27.3 Signature BytesAll Atmel microcontrollers have a three-byte signature code which identifies the device. This code can be read inboth serial and parallel mode, also when the device is locked. The three bytes reside in a separate address space.

For the ATmega32A the signature bytes are:

1. $000: $1E (indicates manufactured by Atmel)

2. $001: $95 (indicates 32Kbytes Flash memory)

3. $002: $02 (indicates ATmega32A device when $001 is $95)

27.4 Calibration ByteThe ATmega32A stores four different calibration values for the internal RC Oscillator. These bytes resides in thesignature row High Byte of the addresses 0x0000, 0x0001, 0x0002, and 0x0003 for 1, 2, 4 , and 8Mhz respectively.During Reset, the 1MHz value is automatically loaded into the OSCCAL Register. If other frequencies are used,the calibration value has to be loaded manually, see “OSCCAL – Oscillator Calibration Register” on page 30“ fordetails.

Table 27-4. Fuse Low Byte

Fuse Low Byte Bit No. Description Default Value

BODLEVEL 7 Brown-out Detector trigger level 1 (unprogrammed)

BODEN 6 Brown-out Detector enable 1 (unprogrammed, BOD disabled)

SUT1 5 Select start-up time 1 (unprogrammed)(1)

SUT0 4 Select start-up time 0 (programmed)(1)

CKSEL3 3 Select Clock source 0 (programmed)(2)



CKSEL0 0 Select Clock source 1 (unprogrammed)(2)


27.5 Page Size

27.6 Parallel Programming Parameters, Pin Mapping, and CommandsThis section describes how to parallel program and verify Flash Program memory, EEPROM Data memory, Mem-ory Lock bits, and Fuse bits in the ATmega32A. Pulses are assumed to be at least 250 ns unless otherwise noted.

27.6.1 Signal NamesIn this section, some pins of the ATmega32A are referenced by signal names describing their functionality duringparallel programming, see Figure 26-1 and Table 26-7. Pins not described in the following table are referenced bypin names.

The XA1/XA0 pins determine the action executed when the XTAL1 pin is given a positive pulse. The bit coding isshown in Table 26-9.

When pulsing WR or OE, the command loaded determines the action executed. The different Commands areshown in Table 26-10.

Figure 27-1. Parallel Programming

Table 27-5. No. of Words in a Page and no. of Pages in the Flash

Flash Size Page Size PCWORD No. of Pages PCPAGE PCMSB

16K words (32Kbytes) 64 words PC[5:0] 256 PC[13:6] 13

Table 27-6. No. of Words in a Page and no. of Pages in the EEPROM

EEPROM Size Page Size PCWORD No. of Pages PCPAGE EEAMSB

1024bytes 4bytes EEA[1:0] 256 EEA[9:2] 9

VCC

+5V

GND

XTAL1

PD1

PD2

PD3

PD4

PD5

PD6

PB7 - PB0 DATA

RESET

PD7

+12 V

BS1

XA0

XA1

OE

RDY/BSY

PAGEL

PA0

WR

BS2

AVCC

+5V


Table 27-7. Pin Name Mapping

Signal Name in Programming Mode

Pin Name I/O Function

RDY/BSY PD1 O0: Device is busy programming, 1: Device is ready for new command

OE PD2 I Output Enable (Active low)

WR PD3 I Write Pulse (Active low)

BS1 PD4 IByte Select 1 (“0” selects low byte, “1” selects high byte)

XA0 PD5 I XTAL Action Bit 0

XA1 PD6 I XTAL Action Bit 1

PAGEL PD7 I Program Memory and EEPROM data Page Load

BS2 PA0 IByte Select 2 (“0” selects low byte, “1” selects 2’nd high byte)

DATA PB7-0 I/O Bidirectional Data bus (Output when OE is low)

Table 27-8. Pin Values used to Enter Programming Mode

Pin Symbol Value

PAGEL Prog_enable[3] 0

XA1 Prog_enable[2] 0

XA0 Prog_enable[1] 0

BS1 Prog_enable[0] 0

Table 27-9. XA1 and XA0 Coding

XA1 XA0 Action when XTAL1 is Pulsed

0 0 Load Flash or EEPROM Address (High or low address byte determined by BS1)

0 1 Load Data (High or Low data byte for Flash determined by BS1)

1 0 Load Command

1 1 No Action, Idle

Table 27-10. Command Byte Bit Coding

Command Byte Command Executed

1000 0000 Chip Erase

0100 0000 Write Fuse Bits

0010 0000 Write Lock Bits

0001 0000 Write Flash

0001 0001 Write EEPROM

0000 1000 Read Signature Bytes and Calibration byte


27.7 Parallel Programming

27.7.1 Enter Programming ModeThe following algorithm puts the device in Parallel Programming mode:

1. Apply 4.5V - 5.5V between VCC and GND, and wait at least 100 µs.

2. Set RESET to “0” and toggle XTAL1 at least 6 times

3. Set the Prog_enable pins listed in Table 26-8 on page 270 to “0000” and wait at least 100 ns.

4. Apply 11.5V - 12.5V to RESET. Any activity on Prog_enable pins within 100 ns after +12V has been applied to RESET, will cause the device to fail entering Programming mode.

Note, if External Crystal or External RC configuration is selected, it may not be possible to apply qualified XTAL1pulses. In such cases, the following algorithm should be followed:

1. Set Prog_enable pins listed in Table 26-8 on page 270 to “0000”.

2. Apply 4.5V - 5.5V between VCC and GND simultanously as 11.5V - 12.5V is applied to RESET.

3. Wait 100 µs.

4. Re-program the fuses to ensure that External Clock is selected as clock source (CKSEL3:0 = 0b0000) If Lock bits are programmed, a Chip Erase command must be executed before changing the fuses.

5. Exit Programming mode by power the device down or by bringing RESET pin to 0b0.

6. Entering Programming mode with the original algorithm, as described above.

27.7.2 Considerations for Efficient ProgrammingThe loaded command and address are retained in the device during programming. For efficient programming, thefollowing should be considered.

• The command needs only be loaded once when writing or reading multiple memory locations.

• Skip writing the data value $FF, that is the contents of the entire EEPROM (unless the EESAVE fuse is programmed) and Flash after a Chip Erase.

• Address high byte needs only be loaded before programming or reading a new 256 word window in Flash or 256 byte EEPROM. This consideration also applies to Signature bytes reading.

27.7.3 Chip EraseThe Chip Erase will erase the Flash and EEPROM(1) memories plus Lock bits. The Lock bits are not reset until theprogram memory has been completely erased. The Fuse bits are not changed. A Chip Erase must be performedbefore the Flash and/or the EEPROM are reprogrammed.

Note: 1. The EEPRPOM memory is preserved during chip erase if the EESAVE Fuse is programmed.

Load Command “Chip Erase”

1. Set XA1, XA0 to “10”. This enables command loading.

2. Set BS1 to “0”.

3. Set DATA to “1000 0000”. This is the command for Chip Erase.

4. Give XTAL1 a positive pulse. This loads the command.

5. Give WR a negative pulse. This starts the Chip Erase. RDY/BSY goes low.

0000 0100 Read Fuse and Lock bits

0000 0010 Read Flash

0000 0011 Read EEPROM

Table 27-10. Command Byte Bit Coding (Continued)

Command Byte Command Executed


6. Wait until RDY/BSY goes high before loading a new command.

27.7.4 Programming the FlashThe Flash is organized in pages, see Table 26-5 on page 269. When programming the Flash, the program data islatched into a page buffer. This allows one page of program data to be programmed simultaneously. The followingprocedure describes how to program the entire Flash memory:

A. Load Command “Write Flash”

1. Set XA1, XA0 to “10”. This enables command loading.

2. Set BS1 to “0”.

3. Set DATA to “0001 0000”. This is the command for Write Flash.

4. Give XTAL1 a positive pulse. This loads the command.

B. Load Address Low byte

1. Set XA1, XA0 to “00”. This enables address loading.

2. Set BS1 to “0”. This selects low address.

3. Set DATA = Address low byte ($00 - $FF).

4. Give XTAL1 a positive pulse. This loads the address low byte.

C. Load Data Low Byte

1. Set XA1, XA0 to “01”. This enables data loading.

2. Set DATA = Data low byte ($00 - $FF).

3. Give XTAL1 a positive pulse. This loads the data byte.

D. Load Data High Byte

1. Set BS1 to “1”. This selects high data byte.

2. Set XA1, XA0 to “01”. This enables data loading.

3. Set DATA = Data high byte ($00 - $FF).

4. Give XTAL1 a positive pulse. This loads the data byte.

E. Latch Data

1. Set BS1 to “1”. This selects high data byte.

2. Give PAGEL a positive pulse. This latches the data bytes. (See Figure 26-3 for signal waveforms)

F. Repeat B through E until the entire buffer is filled or until all data within the page is loaded.

While the lower bits in the address are mapped to words within the page, the higher bits address the pages withinthe FLASH. This is illustrated in Figure 26-2 on page 273. Note that if less than 8 bits are required to addresswords in the page (page size < 256), the most significant bit(s) in the address low byte are used to address thepage when performing a page write.

G. Load Address High byte

1. Set XA1, XA0 to “00”. This enables address loading.

2. Set BS1 to “1”. This selects high address.

3. Set DATA = Address high byte ($00 - $FF).

4. Give XTAL1 a positive pulse. This loads the address high byte.

H. Program Page

1. Set BS1 = “0”

2. Give WR a negative pulse. This starts programming of the entire page of data. RDY/BSYgoes low.

3. Wait until RDY/BSY goes high. (See Figure 26-3 for signal waveforms)

I. Repeat B through H until the entire Flash is programmed or until all data has been programmed.


J. End Page Programming

1. 1. Set XA1, XA0 to “10”. This enables command loading.

2. Set DATA to “0000 0000”. This is the command for No Operation.

3. Give XTAL1 a positive pulse. This loads the command, and the internal write signals are reset.

Figure 27-2. Addressing the Flash which is Organized in Pages

Note: 1. PCPAGE and PCWORD are listed in Table 26-5 on page 269.

Figure 27-3. Programming the Flash Waveforms(1)

Note: 1. “XX” is don’t care. The letters refer to the programming description above.

PROGRAM MEMORY

WORD ADDRESSWITHIN A PAGE

PAGE ADDRESSWITHIN THE FLASH

INSTRUCTION WORD

PAGE PCWORD[PAGEMSB:0]:

00

01

02

PAGEEND

PAGE

PCWORDPCPAGE

PCMSB PAGEMSBPROGRAMCOUNTER

RDY/BSY

WR

OE

RESET +12V

PAGEL

BS2

$10 ADDR. LOW ADDR. HIGHDATADATA LOW DATA HIGH ADDR. LOW DATA LOW DATA HIGH

XA1

XA0

BS1

XTAL1

XX XX XX

A B C D E B C D E G H

F


27.7.5 Programming the EEPROMThe EEPROM is organized in pages, see Table 26-6 on page 269. When programming the EEPROM, the programdata is latched into a page buffer. This allows one page of data to be programmed simultaneously. The program-ming algorithm for the EEPROM data memory is as follows (refer to “Programming the Flash” on page 272 fordetails on Command, Address and Data loading):

1. A: Load Command “0001 0001”.

2. G: Load Address High Byte ($00 - $FF)

3. B: Load Address Low Byte ($00 - $FF)

4. C: Load Data ($00 - $FF)

5. E: Latch data (give PAGEL a positive pulse)

K: Repeat 3 through 5 until the entire buffer is filled

L: Program EEPROM page

1. Set BS1 to “0”.

2. Give WR a negative pulse. This starts programming of the EEPROM page. RDY/BSY goes low.

3. Wait until to RDY/BSY goes high before programming the next page. (See Figure 26-4 for signal waveforms)

Figure 27-4. Programming the EEPROM Waveforms

27.7.6 Reading the FlashThe algorithm for reading the Flash memory is as follows (refer to “Programming the Flash” on page 272 for detailson Command and Address loading):

1. A: Load Command “0000 0010”.



4. Set OE to “0”, and BS1 to “0”. The Flash word low byte can now be read at DATA.

5. Set BS1 to “1”. The Flash word high byte can now be read at DATA.

RDY/BSY

WR

OE

RESET +12V

PAGEL

BS2

0x11 ADDR. HIGHDATA

ADDR. LOW DATA ADDR. LOW DATA XX

XA1

XA0

BS1

XTAL1

XX

A G B C E B C E L

K


6. Set OE to “1”.

27.7.7 Reading the EEPROMThe algorithm for reading the EEPROM memory is as follows (refer to “Programming the Flash” on page 272 fordetails on Command and Address loading):

1. A: Load Command “0000 0011”.



4. Set OE to “0”, and BS1 to “0”. The EEPROM Data byte can now be read at DATA.


27.7.8 Programming the Fuse Low BitsThe algorithm for programming the Fuse Low bits is as follows (refer to “Programming the Flash” on page 272 fordetails on Command and Data loading):

1. A: Load Command “0100 0000”.

2. C: Load Data Low Byte. Bit n = “0” programs and bit n = “1” erases the Fuse bit.

3. Set BS1 to “0” and BS2 to “0”.

4. Give WR a negative pulse and wait for RDY/BSY to go high.

27.7.9 Programming the Fuse High BitsThe algorithm for programming the Fuse high bits is as follows (refer to “Programming the Flash” on page 272 fordetails on Command and Data loading):

1. A: Load Command “0100 0000”.

2. C: Load Data Low Byte. Bit n = “0” programs and bit n = “1” erases the Fuse bit.

3. Set BS1 to “1” and BS2 to “0”. This selects high data byte.


5. Set BS1 to “0”. This selects low data byte.


Figure 27-5. Programming the Fuses

27.7.10 Programming the Lock BitsThe algorithm for programming the Lock bits is as follows (refer to “Programming the Flash” on page 272 for detailson Command and Data loading):

1. A: Load Command “0010 0000”.

2. C: Load Data Low Byte. Bit n = “0” programs the Lock bit.


The Lock bits can only be cleared by executing Chip Erase.

27.7.11 Reading the Fuse and Lock BitsThe algorithm for reading the Fuse and Lock bits is as follows (refer to “Programming the Flash” on page 272 fordetails on Command loading):

1. A: Load Command “0000 0100”.

2. Set OE to “0”, BS2 to “0” and BS1 to “0”. The status of the Fuse Low bits can now be read at DATA (“0” means programmed).

3. Set OE to “0”, BS2 to “1” and BS1 to “1”. The status of the Fuse High bits can now be read at DATA (“0” means programmed).

4. Set OE to “0”, BS2 to “0” and BS1 to “1”. The status of the Lock bits can now be read at DATA (“0” means programmed).


RDY/BSY

WR

OE

RESET +12V

PAGEL

$40DATA

DATA XX

XA1

XA0

BS1

XTAL1

A C

$40 DATA XX

A C

Write Fuse Low byte Write Fuse high byte

BS2


Figure 27-6. Mapping between BS1, BS2 and the Fuse- and Lock Bits during Read

27.7.12 Reading the Signature BytesThe algorithm for reading the Signature bytes is as follows (refer to “Programming the Flash” on page 272 fordetails on Command and Address loading):

1. A: Load Command “0000 1000”.

2. B: Load Address Low Byte ($00 - $02).

3. Set OE to “0”, and BS1 to “0”. The selected Signature byte can now be read at DATA.


27.7.13 Reading the Calibration ByteThe algorithm for reading the Calibration byte is as follows (refer to “Programming the Flash” on page 272 fordetails on Command and Address loading):

1. A: Load Command “0000 1000”.

2. B: Load Address Low Byte, $00.

3. Set OE to “0”, and BS1 to “1”. The Calibration byte can now be read at DATA.


27.7.14 Parallel Programming Characteristics

Figure 27-7. Parallel Programming Timing, Including some General Timing Requirements

Fuse Low Byte

Lock Bits 0

1

BS2

Fuse High Byte

0

1

BS1

DATA

Data & Contol(DATA, XA0/1, BS1, BS2)

XTAL1tXHXL

tWL WH

tDVXH tXLDX

tPLWL

tWLRH

WR

RDY/BSY

PAGEL tPHPL

tPLBXtBVPH

tXLWL

tWLBXt BVWL

WLRL


Figure 27-8. Parallel Programming Timing, Loading Sequence with Timing Requirements(1)

Note: 1. The timing requirements shown in Figure 26-7 (that is, tDVXH, tXHXL, and tXLDX) also apply to loading operation.

Figure 27-9. Parallel Programming Timing, Reading Sequence (within the Same Page) with Timing Requirements(1)

Note: 1. The timing requirements shown in Figure 26-7 (that is, tDVXH, tXHXL, and tXLDX) also apply to reading operation.

Table 27-11. Parallel Programming Characteristics, VCC = 5V ±10%

Symbol Parameter Min Typ Max Units

VPP Programming Enable Voltage 11.5 12.5 V

IPP Programming Enable Current 250 A

tDVXH Data and Control Valid before XTAL1 High 67 ns

tXLXH XTAL1 Low to XTAL1 High 200 ns

tXHXL XTAL1 Pulse Width High 150 ns

tXLDX Data and Control Hold after XTAL1 Low 67 ns

tXLWL XTAL1 Low to WR Low 0 ns

tXLPH XTAL1 Low to PAGEL high 0 ns

tPLXH PAGEL low to XTAL1 high 150 ns

tBVPH BS1 Valid before PAGEL High 67 ns

XTAL1

PAGEL

tPLXHXLXHt tXLPH

ADDR0 (Low Byte) DATA (Low Byte) DATA (High Byte) ADDR1 (Low Byte)DATA

BS1

XA0

XA1

LOAD ADDRESS(LOW BYTE)

LOAD DATA (LOW BYTE)

LOAD DATA(HIGH BYTE)

LOAD DATA LOAD ADDRESS(LOW BYTE)

XTAL1

OE

ADDR0 (Low Byte) DATA (Low Byte) DATA (High Byte) ADDR1 (Low Byte)DATA

BS1

XA0

XA1


READ DATA (LOW BYTE)

READ DATA(HIGH BYTE)


tBVDV

tOLDV

tXLOL

tOHDZ


Notes: 1. tWLRH is valid for the Write Flash, Write EEPROM, Write Fuse bits and Write Lock bits commands.

2. tWLRH_CE is valid for the Chip Erase command.

27.8 SPI Serial DownloadingBoth the Flash and EEPROM memory arrays can be programmed using the serial SPI bus while RESET is pulledto GND. The serial interface consists of pins SCK, MOSI (input), and MISO (output). After RESET is set low, theProgramming Enable instruction needs to be executed first before program/erase operations can be executed.NOTE, in Table 26-12 on page 280, the pin mapping for SPI programming is listed. Not all parts use the SPI pinsdedicated for the internal SPI interface.

27.9 SPI Serial Programming Pin Mapping

tPHPL PAGEL Pulse Width High 150 ns

tPLBX BS1 Hold after PAGEL Low 67 ns

tWLBX BS2/1 Hold after WR Low 67 ns

tPLWL PAGEL Low to WR Low 67 ns

tBVWL BS1 Valid to WR Low 67 ns

tWLWH WR Pulse Width Low 150 ns

tWLRL WR Low to RDY/BSY Low 0 1 s

tWLRH WR Low to RDY/BSY High(1) 3.7 4.5 ms

tWLRH_CE WR Low to RDY/BSY High for Chip Erase(2) 7.5 9 ms

tXLOL XTAL1 Low to OE Low 0 ns

tBVDV BS1 Valid to DATA valid 0 250 ns

tOLDV OE Low to DATA Valid 250 ns

tOHDZ OE High to DATA Tri-stated 250 ns

Table 27-11. Parallel Programming Characteristics, VCC = 5V ±10% (Continued)

Symbol Parameter Min Typ Max Units

Table 27-12. Pin Mapping SPI Serial Programming

Symbol Pins I/O Description

MOSI PB5 I Serial Data in

MISO PB6 O Serial Data out

SCK PB7 I Serial Clock


Figure 27-10. SPI Serial Programming and Verify(1)

Notes: 1. If the device is clocked by the Internal Oscillator, it is no need to connect a clock source to the XTAL1 pin.

2. VCC -0.3V < AVCC < VCC +0.3V, however, AVCC should always be within 2.7V - 5.5V

When programming the EEPROM, an auto-erase cycle is built into the self-timed programming operation (in theserial mode ONLY) and there is no need to first execute the Chip Erase instruction. The Chip Erase operation turnsthe content of every memory location in both the Program and EEPROM arrays into $FF.

Depending on CKSEL Fuses, a valid clock must be present. The minimum low and high periods for the serial clock(SCK) input are defined as follows:

Low: > 2 CPU clock cycles for fck < 12MHz, 3 CPU clock cycles for fck 12MHz

High: > 2 CPU clock cycles for fck < 12MHz, 3 CPU clock cycles for fck 12MHz

27.9.1 SPI Serial Programming AlgorithmWhen writing serial data to the ATmega32A, data is clocked on the rising edge of SCK.

When reading data from the ATmega32A, data is clocked on the falling edge of SCK. See Figure 26-11 for timingdetails.

To program and verify the ATmega32A in the SPI Serial Programming mode, the following sequence is recom-mended (See four byte instruction formats in Table 26-14):

1. Power-up sequence:Apply power between VCC and GND while RESET and SCK are set to “0”. In some systems, the program-mer can not guarantee that SCK is held low during power-up. In this case, RESET must be given a positive pulse of at least two CPU clock cycles duration after SCK has been set to “0”.

2. Wait for at least 20ms and enable SPI Serial Programming by sending the Programming Enable serial instruction to pin MOSI.

3. The SPI Serial Programming instructions will not work if the communication is out of synchronization. When in sync. the second byte ($53), will echo back when issuing the third byte of the Programming Enable instruction. Whether the echo is correct or not, all four bytes of the instruction must be transmitted. If the $53 did not echo back, give RESET a positive pulse and issue a new Programming Enable command.

4. The Flash is programmed one page at a time (page size found in “Page Size” on page 269). The memory page is loaded one byte at a time by supplying the 6 LSB of the address and data together with the Load Program Memory Page instruction. To ensure correct loading of the page, the data low byte must be

VCC

GND

XTAL1

SCK

MISO

MOSI

RESET

PB5

PB6

PB7

+2.7 - 5.5V

AVCC

+2.7 - 5.5V(2)


loaded before data high byte is applied for a given address. The Program Memory Page is stored by load-ing the Write Program Memory Page instruction with the 8 MSB of the address. If polling is not used, the user must wait at least tWD_FLASH before issuing the next page. (See Table 26-13). Accessing the SPI Serial Programming interface before the Flash write operation completes can result in incorrect programming.

5. The EEPROM array is programmed one byte at a time by supplying the address and data together with the appropriate Write instruction. An EEPROM memory location is first automatically erased before new data is written. If polling is not used, the user must wait at least tWD_EEPROM before issuing the next byte. (See Table 26-13). In a chip erased device, no $FFs in the data file(s) need to be programmed.

6. Any memory location can be verified by using the Read instruction which returns the content at the selected address at serial output MISO.

7. At the end of the programming session, RESET can be set high to commence normal operation.

8. Power-off sequence (if needed):Set RESET to “1”.Turn VCC power off.

27.9.2 Data Polling FlashWhen a page is being programmed into the Flash, reading an address location within the page being programmedwill give the value $FF. At the time the device is ready for a new page, the programmed value will read correctly.This is used to determine when the next page can be written. Note that the entire page is written simultaneouslyand any address within the page can be used for polling. Data polling of the Flash will not work for the value $FF,so when programming this value, the user will have to wait for at least tWD_FLASH before programming the nextpage. As a chip erased device contains $FF in all locations, programming of addresses that are meant to contain$FF, can be skipped. See Table 26-13 for tWD_FLASH value

27.9.3 Data Polling EEPROMWhen a new byte has been written and is being programmed into EEPROM, reading the address location beingprogrammed will give the value $FF. At the time the device is ready for a new byte, the programmed value will readcorrectly. This is used to determine when the next byte can be written. This will not work for the value $FF, but theuser should have the following in mind: As a chip erased device contains $FF in all locations, programming ofaddresses that are meant to contain $FF, can be skipped. This does not apply if the EEPROM is re-programmedwithout chip erasing the device. In this case, data polling cannot be used for the value $FF, and the user will haveto wait at least tWD_EEPROM before programming the next byte. See Table 26-13 for tWD_EEPROM value.

Table 27-13. Minimum Wait Delay before Writing the Next Flash or EEPROM Location

Symbol Minimum Wait Delay

tWD_FLASH 4.5ms

tWD_EEPROM 9.0ms

tWD_ERASE 9.0ms

tWD_FUSE 4.5ms


Figure 27-11. SPI Serial Programming Waveforms

MSB

MSB

LSB

LSB

SERIAL CLOCK INPUT(SCK)

SERIAL DATA INPUT (MOSI)

(MISO)

SAMPLE

SERIAL DATA OUTPUT


Note: a = address high bitsb = address low bitsH = 0 – Low byte, 1 – High Byteo = data outi = data inx = don’t care

27.9.4 SPI Serial Programming CharacteristicsFor Characteristics of SPI module, see “SPI Timing Characteristics” on page 301.

Table 27-14. SPI Serial Programming Instruction Set

Instruction

Instruction Format

OperationByte 1 Byte 2 Byte 3 Byte4

Programming Enable1010 1100 0101 0011 xxxx xxxx xxxx xxxx Enable SPI Serial Programming after

RESET goes low.

Chip Erase 1010 1100 100x xxxx xxxx xxxx xxxx xxxx Chip Erase EEPROM and Flash.

Read Program Memory0010 H000 00aa aaaa bbbb bbbb oooo oooo Read H (high or low) data o from

Program memory at word address a:b.

Load Program Memory Page

0100 H000 00xx xxxx xxbb bbbb iiii iiii Write H (high or low) data i to Program Memory page at word address b. Data low byte must be loaded before Data high byte is applied within the same address.

Write Program Memory Page0100 1100 00aa aaaa bbxx xxxx xxxx xxxx Write Program Memory Page at

address a:b.

Read EEPROM Memory1010 0000 00xx xxaa bbbb bbbb oooo oooo Read data o from EEPROM memory at

address a:b.

Write EEPROM Memory1100 0000 00xx xxaa bbbb bbbb iiii iiii Write data i to EEPROM memory at

address a:b.

Read Lock Bits0101 1000 0000 0000 xxxx xxxx xxoo oooo Read Lock bits. “0” = programmed, “1”

= unprogrammed. See Table 26-1 on page 266 for details.

Write Lock Bits1010 1100 111x xxxx xxxx xxxx 11ii iiii Write Lock bits. Set bits = “0” to

program Lock bits. See Table 26-1 on page 266 for details.

Read Signature Byte 0011 0000 00xx xxxx xxxx xxbb oooo oooo Read Signature Byte o at address b.

Write Fuse Bits1010 1100 1010 0000 xxxx xxxx iiii iiii Set bits = “0” to program, “1” to

unprogram. See Table 26-4 on page 268 for details.

Write Fuse High Bits1010 1100 1010 1000 xxxx xxxx iiii iiii Set bits = “0” to program, “1” to

unprogram. See Table 26-3 on page 267 for details.

Read Fuse Bits0101 0000 0000 0000 xxxx xxxx oooo oooo Read Fuse bits. “0” = programmed, “1”

= unprogrammed. See Table 26-4 on page 268 for details.

Read Fuse High Bits0101 1000 0000 1000 xxxx xxxx oooo oooo Read Fuse high bits. “0” = pro-

grammed, “1” = unprogrammed. See Table 26-3 on page 267 for details.

Read Calibration Byte 0011 1000 xxxx xxxx 0000 00bb oooo oooo Read Calibration Byte o at address b


27.10 Programming via the JTAG InterfaceProgramming through the JTAG interface requires control of the four JTAG specific pins: TCK, TMS, TDI and TDO.Control of the reset and clock pins is not required.

To be able to use the JTAG interface, the JTAGEN Fuse must be programmed. The device is default shipped withthe fuse programmed. In addition, the JTD bit in MCUCSR must be cleared. Alternatively, if the JTD bit is set, theExternal Reset can be forced low. Then, the JTD bit will be cleared after two chip clocks, and the JTAG pins areavailable for programming. This provides a means of using the JTAG pins as normal port pins in running modewhile still allowing In-System Programming via the JTAG interface. Note that this technique can not be used whenusing the JTAG pins for Boundary-scan or On-chip Debug. In these cases the JTAG pins must be dedicated forthis purpose.

As a definition in this datasheet, the LSB is shifted in and out first of all Shift Registers.

27.10.1 Programming Specific JTAG InstructionsThe instruction register is 4-bit wide, supporting up to 16 instructions. The JTAG instructions useful for Program-ming are listed below.

The OPCODE for each instruction is shown behind the instruction name in hex format. The text describes whichData Register is selected as path between TDI and TDO for each instruction.

The Run-Test/Idle state of the TAP controller is used to generate internal clocks. It can also be used as an idlestate between JTAG sequences. The state machine sequence for changing the instruction word is shown in Figure26-12.


Figure 27-12. State Machine Sequence for Changing the Instruction Word

27.10.2 AVR_RESET ($C)The AVR specific public JTAG instruction for setting the AVR device in the Reset mode or taking the device outfrom the Reset Mode. The TAP controller is not reset by this instruction. The one bit Reset Register is selected asData Register. Note that the Reset will be active as long as there is a logic “one” in the Reset Chain. The outputfrom this chain is not latched.


• Shift-DR: The Reset Register is shifted by the TCK input.

27.10.3 PROG_ENABLE ($4)The AVR specific public JTAG instruction for enabling programming via the JTAG port. The 16-bit ProgrammingEnable Register is selected as Data Register. The active states are the following:

• Shift-DR: The programming enable signature is shifted into the Data Register.

• Update-DR: The programming enable signature is compared to the correct value, and Programming mode is entered if the signature is valid.

Test-Logic-Reset

Run-Test/Idle

Shift-DR

Exit1-DR

Pause-DR

Exit2-DR

Update-DR

Select-IR Scan

Capture-IR

Shift-IR

Exit1-IR

Pause-IR

Exit2-IR

Update-IR

Select-DR Scan

Capture-DR

0

1

0 1 1 1

0 0

0 0

1 1

1 0

1

1

0

1

0

0

1 0

1

1

0

1

0

0

00

11


27.10.4 PROG_COMMANDS ($5)The AVR specific public JTAG instruction for entering programming commands via the JTAG port. The 15-bit Pro-gramming Command Register is selected as Data Register. The active states are the following:

• Capture-DR: The result of the previous command is loaded into the Data Register.

• Shift-DR: The Data Register is shifted by the TCK input, shifting out the result of the previous command and shifting in the new command.

• Update-DR: The programming command is applied to the Flash inputs

• Run-Test/Idle: One clock cycle is generated, executing the applied command (not always required, see Table 26-15 below).

27.10.5 PROG_PAGELOAD ($6)The AVR specific public JTAG instruction to directly load the Flash data page via the JTAG port. The 1024 bit Vir-tual Flash Page Load Register is selected as Data Register. This is a virtual scan chain with length equal to thenumber of bits in one Flash page. Internally the Shift Register is 8-bit. Unlike most JTAG instructions, the Update-DR state is not used to transfer data from the Shift Register. The data are automatically transferred to the Flashpage buffer byte by byte in the Shift-DR state by an internal state machine. This is the only active state:

• Shift-DR: Flash page data are shifted in from TDI by the TCK input, and automatically loaded into the Flash page one byte at a time.

Note: The JTAG instruction PROG_PAGELOAD can only be used if the AVR device is the first device in JTAG scan chain. If the AVR cannot be the first device in the scan chain, the byte-wise programming algorithm must be used.

27.10.6 PROG_PAGEREAD ($7)The AVR specific public JTAG instruction to read one full Flash data page via the JTAG port. The 1032 bit VirtualFlash Page Read Register is selected as Data Register. This is a virtual scan chain with length equal to the numberof bits in one Flash page plus 8. Internally the Shift Register is 8-bit. Unlike most JTAG instructions, the Capture-DR state is not used to transfer data to the Shift Register. The data are automatically transferred from the Flashpage buffer byte by byte in the Shift-DR state by an internal state machine. This is the only active state:

• Shift-DR: Flash data are automatically read one byte at a time and shifted out on TDO by the TCK input. The TDI input is ignored.

Note: The JTAG instruction PROG_PAGEREAD can only be used if the AVR device is the first device in JTAG scan chain. If the AVR cannot be the first device in the scan chain, the byte-wise programming algorithm must be used.

27.10.7 Data RegistersThe Data Registers are selected by the JTAG Instruction Registers described in section “Programming SpecificJTAG Instructions” on page 284. The Data Registers relevant for programming operations are:

• Reset Register

• Programming Enable Register

• Programming Command Register

• Virtual Flash Page Load Register

• Virtual Flash Page Read Register

27.10.8 Reset RegisterThe Reset Register is a Test Data Register used to reset the part during programming. It is required to reset thepart before entering programming mode.

A high value in the Reset Register corresponds to pulling the external Reset low. The part is reset as long as thereis a high value present in the Reset Register. Depending on the Fuse settings for the clock options, the part willremain reset for a Reset Time-out Period (refer to “Clock Sources” on page 26) after releasing the Reset Register.


The output from this Data Register is not latched, so the reset will take place immediately, as shown in Figure 24-2on page 235.

27.10.9 Programming Enable RegisterThe Programming Enable Register is a 16-bit register. The contents of this register is compared to the program-ming enable signature, binary code 1010_0011_0111_0000. When the contents of the register is equal to theprogramming enable signature, programming via the JTAG port is enabled. The register is reset to 0 on Power-onReset, and should always be reset when leaving Programming mode.

Figure 27-13. Programming Enable Register

27.10.10 Programming Command RegisterThe Programming Command Register is a 15-bit register. This register is used to serially shift in programmingcommands, and to serially shift out the result of the previous command, if any. The JTAG Programming InstructionSet is shown in Table 26-15. The state sequence when shifting in the programming commands is illustrated in Fig-ure 26-16.

TDI

TDO

DATA

= D Q

ClockDR & PROG_ENABLE

Programming Enable$A370


Figure 27-14. Programming Command RegisterTDI

TDO

STROBES

ADDRESS/DATA

FlashEEPROM

FusesLock Bits

Figure 27-15. JTAG Programming Instruction Set a = address high bits, b = address low bits, H = 0 – Low byte, 1 – High Byte, o = data out, i = data in, x = don’t care

Instruction TDI sequence TDO sequence Notes

1a. Chip erase 0100011_10000000

0110001_10000000

0110011_100000000110011_10000000

xxxxxxx_xxxxxxxx

xxxxxxx_xxxxxxxx

xxxxxxx_xxxxxxxxxxxxxxx_xxxxxxxx

1b. Poll for chip erase complete 0110011_10000000 xxxxxox_xxxxxxxx (2)

2a. Enter Flash Write 0100011_00010000 xxxxxxx_xxxxxxxx

2b. Load Address High Byte 0000111_aaaaaaaa xxxxxxx_xxxxxxxx (9)

2c. Load Address Low Byte 0000011_bbbbbbbb xxxxxxx_xxxxxxxx

2d. Load Data Low Byte 0010011_iiiiiiii xxxxxxx_xxxxxxxx

2e. Load Data High Byte 0010111_iiiiiiii xxxxxxx_xxxxxxxx

2f. Latch Data 0110111_00000000

1110111_000000000110111_00000000

xxxxxxx_xxxxxxxx


(1)

2g. Write Flash Page 0110111_000000000110101_00000000

0110111_00000000

0110111_00000000


xxxxxxx_xxxxxxxx

xxxxxxx_xxxxxxxx

(1)

2h. Poll for Page Write complete 0110111_00000000 xxxxxox_xxxxxxxx (2)

3a. Enter Flash Read 0100011_00000010 xxxxxxx_xxxxxxxx




3d. Read Data Low and High Byte 0110010_00000000

0110110_00000000

0110111_00000000

xxxxxxx_xxxxxxxx

xxxxxxx_ooooooooxxxxxxx_oooooooo

low byte

high byte

4a. Enter EEPROM Write 0100011_00010001 xxxxxxx_xxxxxxxx



4d. Load Data Byte 0010011_iiiiiiii xxxxxxx_xxxxxxxx

4e. Latch Data 0110111_000000001110111_00000000

0110111_00000000


xxxxxxx_xxxxxxxx

(1)

4f. Write EEPROM Page 0110011_00000000

0110001_00000000

0110011_000000000110011_00000000

xxxxxxx_xxxxxxxx

xxxxxxx_xxxxxxxx


(1)

4g. Poll for Page Write complete 0110011_00000000 xxxxxox_xxxxxxxx (2)

5a. Enter EEPROM Read 0100011_00000011 xxxxxxx_xxxxxxxx



5d. Read Data Byte 0110011_bbbbbbbb0110010_000000000110011_00000000

xxxxxxx_xxxxxxxx

xxxxxxx_xxxxxxxxxxxxxxx_oooooooo

6a. Enter Fuse Write 0100011_01000000 xxxxxxx_xxxxxxxx

6b. Load Data Low Byte(6) 0010011_iiiiiiii xxxxxxx_xxxxxxxx (3)

6c. Write Fuse High byte 0110111_00000000

0110101_00000000

0110111_000000000110111_00000000

xxxxxxx_xxxxxxxx

xxxxxxx_xxxxxxxx


(1)

6d. Poll for Fuse Write complete 0110111_00000000 xxxxxox_xxxxxxxx (2)

6e. Load Data Low Byte(7) 0010011_iiiiiiii xxxxxxx_xxxxxxxx (3)

6f. Write Fuse Low byte 0110011_00000000

0110001_00000000

0110011_000000000110011_00000000

xxxxxxx_xxxxxxxx

xxxxxxx_xxxxxxxx


(1)

6g. Poll for Fuse Write complete 0110011_00000000 xxxxxox_xxxxxxxx (2)

7a. Enter Lock Bit Write 0100011_00100000 xxxxxxx_xxxxxxxx

7b. Load Data Byte(8) 0010011_11iiiiii xxxxxxx_xxxxxxxx (4)

Figure 27-15. JTAG Programming Instruction Set (Continued) a = address high bits, b = address low bits, H = 0 – Low byte, 1 – High Byte, o = data out, i = data in, x = don’t care



Notes: 1. This command sequence is not required if the seven MSB are correctly set by the previous command sequence (which is normally the case).

2. Repeat until o = “1”.

3. Set bits to “0” to program the corresponding fuse, “1” to unprogram the fuse.

4. Set bits to “0” to program the corresponding lock bit, “1” to leave the lock bit unchanged.

5. “0” = programmed, “1” = unprogrammed.

6. The bit mapping for fuses high byte is listed in Table 26-3 on page 267

7. The bit mapping for fuses low byte is listed in Table 26-4 on page 268

8. The bit mapping for Lock bits byte is listed in Table 26-1 on page 266

9. Address bits exceeding PCMSB and EEAMSB (Table 26-5 and Table 26-6) are don’t care

7c. Write Lock Bits 0110011_00000000

0110001_000000000110011_00000000

0110011_00000000

xxxxxxx_xxxxxxxx


xxxxxxx_xxxxxxxx

(1)

7d. Poll for Lock Bit Write complete 0110011_00000000 xxxxxox_xxxxxxxx (2)

8a. Enter Fuse/Lock Bit Read 0100011_00000100 xxxxxxx_xxxxxxxx

8b. Read Fuse High Byte(6) 0111110_00000000

0111111_00000000

xxxxxxx_xxxxxxxx

xxxxxxx_oooooooo

8c. Read Fuse Low Byte(7) 0110010_00000000

0110011_00000000

xxxxxxx_xxxxxxxx

xxxxxxx_oooooooo

8d. Read Lock Bits(8) 0110110_00000000

0110111_00000000

xxxxxxx_xxxxxxxx

xxxxxxx_xxoooooo(5)

8e. Read Fuses and Lock Bits 0111110_00000000

0110010_00000000

0110110_000000000110111_00000000

xxxxxxx_xxxxxxxx

xxxxxxx_ooooooooxxxxxxx_ooooooooxxxxxxx_oooooooo

(5)

fuse high byte

fuse low bytelock bits

9a. Enter Signature Byte Read 0100011_00001000 xxxxxxx_xxxxxxxx

9b. Load Address Byte 0000011_bbbbbbbb xxxxxxx_xxxxxxxx

9c. Read Signature Byte 0110010_00000000

0110011_00000000

xxxxxxx_xxxxxxxx

xxxxxxx_oooooooo

10a. Enter Calibration Byte Read 0100011_00001000 xxxxxxx_xxxxxxxx

10b. Load Address Byte 0000011_bbbbbbbb xxxxxxx_xxxxxxxx

10c. Read Calibration Byte 0110110_00000000

0110111_00000000

xxxxxxx_xxxxxxxx

xxxxxxx_oooooooo

11a. Load No Operation Command 0100011_00000000

0110011_00000000

xxxxxxx_xxxxxxxx

xxxxxxx_xxxxxxxx

Figure 27-15. JTAG Programming Instruction Set (Continued) a = address high bits, b = address low bits, H = 0 – Low byte, 1 – High Byte, o = data out, i = data in, x = don’t care



Figure 27-16. State Machine Sequence for Changing/Reading the Data Word

27.10.11 Virtual Flash Page Load RegisterThe Virtual Flash Page Load Register is a virtual scan chain with length equal to the number of bits in one Flashpage. Internally the Shift Register is 8-bit, and the data are automatically transferred to the Flash page buffer byteby byte. Shift in all instruction words in the page, starting with the LSB of the first instruction in the page and endingwith the MSB of the last instruction in the page. This provides an efficient way to load the entire Flash page bufferbefore executing Page Write.

Test-Logic-Reset

Run-Test/Idle

Shift-DR

Exit1-DR

Pause-DR

Exit2-DR

Update-DR

Select-IR Scan

Capture-IR

Shift-IR

Exit1-IR

Pause-IR

Exit2-IR

Update-IR

Select-DR Scan

Capture-DR

0

1

0 1 1 1

0 0

0 0

1 1

1 0

1

1

0

1

0

0

1 0

1

1

0

1

0

0

00

11


Figure 27-17. Virtual Flash Page Load Register

27.10.12 Virtual Flash Page Read RegisterThe Virtual Flash Page Read Register is a virtual scan chain with length equal to the number of bits in one Flashpage plus 8. Internally the Shift Register is 8-bit, and the data are automatically transferred from the Flash datapage byte by byte. The first 8 cycles are used to transfer the first byte to the internal Shift Register, and the bits thatare shifted out during these 8 cycles should be ignored. Following this initialization, data are shifted out startingwith the LSB of the first instruction in the page and ending with the MSB of the last instruction in the page. This pro-vides an efficient way to read one full Flash page to verify programming.

Figure 27-18. Virtual Flash Page Read Register

27.10.13 Programming AlgorithmAll references below of type “1a”, “1b”, and so on, refer to Table 26-15.

TDI

TDO

DATA

FlashEEPROM

FusesLock Bits

STROBES

ADDRESS

StateMachine

TDI

TDO

DATA

FlashEEPROM

FusesLock Bits

STROBES

ADDRESS

StateMachine


27.10.14 Entering Programming Mode

1. Enter JTAG instruction AVR_RESET and shift 1 in the Reset Register.

2. Enter instruction PROG_ENABLE and shift 1010_0011_0111_0000 in the Programming Enable Register.

27.10.15 Leaving Programming Mode

1. Enter JTAG instruction PROG_COMMANDS.

2. Disable all programming instructions by usning no operation instruction 11a.

3. Enter instruction PROG_ENABLE and shift 0000_0000_0000_0000 in the programming Enable Register.

4. Enter JTAG instruction AVR_RESET and shift 0 in the Reset Register.

27.10.16 Performing Chip Erase


2. Start chip erase using programming instruction 1a.

3. Poll for Chip Erase complete using programming instruction 1b, or wait for tWLRH_CE (refer to Table 26-11 on page 279).

27.10.17 Programming the FlashBefore programming the Flash a Chip Erase must be performed. See “Performing Chip Erase” on page 293.


2. Enable Flash write using programming instruction 2a.

3. Load address high byte using programming instruction 2b.

4. Load address low byte using programming instruction 2c.

5. Load data using programming instructions 2d, 2e and 2f.

6. Repeat steps 4 and 5 for all instruction words in the page.

7. Write the page using programming instruction 2g.

8. Poll for Flash write complete using programming instruction 2h, or wait for tWLRH (refer to Table 26-11 on page 279).

9. Repeat steps 3 to 7 until all data have been programmed.

A more efficient data transfer can be achieved using the PROG_PAGELOAD instruction:


2. Enable Flash write using programming instruction 2a.

3. Load the page address using programming instructions 2b and 2c. PCWORD (refer to Table 26-5 on page 269) is used to address within one page and must be written as 0.

4. Enter JTAG instruction PROG_PAGELOAD.

5. Load the entire page by shifting in all instruction words in the page, starting with the LSB of the first instruction in the page and ending with the MSB of the last instruction in the page.


7. Write the page using programming instruction 2g.

8. Poll for Flash write complete using programming instruction 2h, or wait for tWLRH (refer to Table 26-11 on page 279).


27.10.18 Reading the Flash


2. Enable Flash read using programming instruction 3a.

3. Load address using programming instructions 3b and 3c.


4. Read data using programming instruction 3d.

5. Repeat steps 3 and 4 until all data have been read.

A more efficient data transfer can be achieved using the PROG_PAGEREAD instruction:


2. Enable Flash read using programming instruction 3a.

3. Load the page address using programming instructions 3b and 3c. PCWORD (refer to Table 26-5 on page 269) is used to address within one page and must be written as 0.

4. Enter JTAG instruction PROG_PAGEREAD.

5. Read the entire page by shifting out all instruction words in the page, starting with the LSB of the first instruction in the page and ending with the MSB of the last instruction in the page. Remember that the first 8 bits shifted out should be ignored.


7. Repeat steps 3 to 6 until all data have been read.

27.10.19 Programming the EEPROMBefore programming the EEPROM a Chip Erase must be performed. See “Performing Chip Erase” on page 293.


2. Enable EEPROM write using programming instruction 4a.

3. Load address high byte using programming instruction 4b.

4. Load address low byte using programming instruction 4c.

5. Load data using programming instructions 4d and 4e.

6. Repeat steps 4 and 5 for all data bytes in the page.

7. Write the data using programming instruction 4f.

8. Poll for EEPROM write complete using programming instruction 4g, or wait for tWLRH (refer to Table 26-11 on page 279).


Note that the PROG_PAGELOAD instruction can not be used when programming the EEPROM

27.10.20 Reading the EEPROM


2. Enable EEPROM read using programming instruction 5a.

3. Load address using programming instructions 5b and 5c.

4. Read data using programming instruction 5d.

5. Repeat steps 3 and 4 until all data have been read.

Note that the PROG_PAGEREAD instruction can not be used when reading the EEPROM

27.10.21 Programming the Fuses


2. Enable Fuse write using programming instruction 6a.

3. Load data high byte using programming instructions 6b. A bit value of “0” will program the corresponding fuse, a “1” will unprogram the fuse.

4. Write Fuse High byte using programming instruction 6c.

5. Poll for Fuse write complete using programming instruction 6d, or wait for tWLRH (refer to Table 26-11 on page 279).

6. Load data low byte using programming instructions 6e. A “0” will program the fuse, a “1” will unprogram the fuse.


7. Write Fuse low byte using programming instruction 6f.

8. Poll for Fuse write complete using programming instruction 6g, or wait for tWLRH (refer to Table 26-11 on page 279).

27.10.22 Programming the Lock Bits


2. Enable Lock bit write using programming instruction 7a.

3. Load data using programming instructions 7b. A bit value of “0” will program the corresponding Lock bit, a “1” will leave the Lock bit unchanged.

4. Write Lock bits using programming instruction 7c.

5. Poll for Lock bit write complete using programming instruction 7d, or wait for tWLRH (refer to Table 26-11 on page 279).

27.10.23 Reading the Fuses and Lock Bits


2. Enable Fuse/Lock bit read using programming instruction 8a.

3. To read all Fuses and Lock bits, use programming instruction 8e.To only read Fuse high byte, use programming instruction 8b.To only read Fuse low byte, use programming instruction 8c.To only read Lock bits, use programming instruction 8d.

27.10.24 Reading the Signature Bytes


2. Enable Signature byte read using programming instruction 9a.

3. Load address $00 using programming instruction 9b.

4. Read first signature byte using programming instruction 9c.

5. Repeat steps 3 and 4 with address $01 and address $02 to read the second and third signature bytes, respectively.

27.10.25 Reading the Calibration Byte


2. Enable Calibration byte read using programming instruction 10a.

3. Load address $00 using programming instruction 10b.

4. Read the calibration byte using programming instruction 10c.


28. Electrical Characteristics

28.1 Absolute Maximum Ratings*

28.2 DC Characteristics

Operating Temperature.................................. -55C to +125C *NOTICE: Stresses beyond those listed under “Absolute Maximum Ratings” may cause permanent dam-age to the device. This is a stress rating only and functional operation of the device at these or other conditions beyond those indicated in the operational sections of this specification is not implied. Exposure to absolute maximum rating conditions for extended periods may affect device reliability.

Storage Temperature ..................................... -65°C to +150°C

Voltage on any Pin except RESETwith respect to Ground ................................-0.5V to VCC+0.5V

Voltage on RESET with respect to Ground......-0.5V to +13.0V

Maximum Operating Voltage ............................................ 6.0V

DC Current per I/O Pin ............................................... 40.0 mA

DC Current VCC and GND Pins......................... 200.0 mA and

400.0 mA TQFP/MLF

TA = -40C to 85C, VCC = 2.7 V to 5.5 V (Unless Otherwise Noted)


VILInput Low Voltage except XTAL1 and RESET pins

VCC = 2.7 - 5.5

VCC = 4.5 - 5.5-0.5 0.2 VCC

(1) V

VIHInput High Voltage except XTAL1 and RESET pins

VCC = 2.7 - 5.5

VCC = 4.5 - 5.50.6 VCC

(2) VCC + 0.5 V

VIL1Input Low Voltage

XTAL1 pinVCC = 2.7 - 5.5 -0.5 0.1 VCC

(1) V

VIH1Input High Voltage

XTAL1 pin

VCC = 2.7 - 5.5

VCC = 4.5 - 5.50.7 VCC

(2) VCC + 0.5 V

VIL2Input Low Voltage

RESET pinVCC = 2.7 - 5.5 -0.5 0.2 VCC V

VIH2Input High Voltage

RESET pinVCC = 2.7 - 5.5 0.9 VCC

(2) VCC + 0.5 V

VOLOutput Low Voltage(3)

(Ports A,B,C,D)IOL = 20 mA, VCC = 5VIOL = 10 mA, VCC = 3V

0.70.5

VV

VOHOutput High Voltage(4)

(Ports A,B,C,D)IOH = -20 mA, VCC = 5VIOH = -10 mA, VCC = 3V

4.22.2

VV

IILInput LeakageCurrent I/O Pin

VCC = 5.5V, pin low(absolute value)

1 µA

IIHInput LeakageCurrent I/O Pin

VCC = 5.5V, pin high(absolute value)

1 µA

RRST Reset Pull-up Resistor 30 60 85 k

Rpu I/O Pin Pull-up Resistor 20 50 k


Notes: 1. “Max” means the highest value where the pin is guaranteed to be read as low

2. “Min” means the lowest value where the pin is guaranteed to be read as high

3. Although each I/O port can sink more than the test conditions (20 mA at Vcc = 5V, 10 mA at Vcc = 3V) under steady state conditions (non-transient), the following must be observed:PDIP Package:1] The sum of all IOL, for all ports, should not exceed 200 mA.2] The sum of all IOL, for port A0 - A7, should not exceed 100 mA.3] The sum of all IOL, for ports B0 - B7,C0 - C7, D0 - D7 and XTAL2, should not exceed 100 mA.TQFP and QFN/MLF Package:1] The sum of all IOL, for all ports, should not exceed 400 mA.2] The sum of all IOL, for ports A0 - A7, should not exceed 100 mA.3] The sum of all IOL, for ports B0 - B4, should not exceed 100 mA.4] The sum of all IOL, for ports B3 - B7, XTAL2, D0 - D2, should not exceed 100 mA.5] The sum of all IOL, for ports D3 - D7, should not exceed 100 mA.6] The sum of all IOL, for ports C0 - C7, should not exceed 100 mA.If IOL exceeds the test condition, VOL may exceed the related specification. Pins are not guaranteed to sink current greater than the listed test condition.

4. Although each I/O port can source more than the test conditions (20 mA at Vcc = 5V, 10 mA at Vcc = 3V) under steady state conditions (non-transient), the following must be observed:PDIP Package:1] The sum of all IOH, for all ports, should not exceed 200 mA.2] The sum of all IOH, for port A0 - A7, should not exceed 100 mA.3] The sum of all IOH, for ports B0 - B7,C0 - C7, D0 - D7 and XTAL2, should not exceed 100 mA.TQFP and QFN/MLF Package:1] The sum of all IOH, for all ports, should not exceed 400 mA.2] The sum of all IOH, for ports A0 - A7, should not exceed 100 mA.3] The sum of all IOH, for ports B0 - B4, should not exceed 100 mA.4] The sum of all IOH, for ports B3 - B7, XTAL2, D0 - D2, should not exceed 100 mA.5] The sum of all IOH, for ports D3 - D7, should not exceed 100 mA.6] The sum of all IOH, for ports C0 - C7, should not exceed 100 mA.If IOH exceeds the test condition, VOH may exceed the related specification. Pins are not guaranteed to source current greater than the listed test condition.

5. Minimum VCC for Power-down is 2.5V.

ICC

Power Supply Current

Active 1MHz, VCC = 3V 0.6 mA

Active 4MHz, VCC = 3V 2.1 5 mA

Active 8MHz, VCC = 5V 7.5 15 mA

Idle 1MHz, VCC = 3V 0.2 mA

Idle 4MHz, VCC = 3V 0.6 2.5 mA

Idle 8MHz, VCC = 5V 2.8 8 mA

Power-down Mode(5)WDT enabled, VCC = 3V < 10 20 µA

WDT disabled, VCC = 3V < 1 10 µA

VACIOAnalog Comparator Input Offset Voltage

VCC = 5V

Vin = VCC/240 mV

IACLKAnalog Comparator Input Leakage Current

VCC = 5VVin = VCC/2

-50 50 nA

tACPDAnalog Comparator Propagation Delay

VCC = 2.7VVCC = 4.0V

750500

ns

TA = -40C to 85C, VCC = 2.7 V to 5.5 V (Unless Otherwise Noted)



28.3 Speed Grades

Figure 28-1. Maximum Frequency vs. VCC.

28.4 Clock Characteristics

28.4.1 External Clock Drive Waveforms

Figure 28-2. External Clock Drive Waveforms

28.4.2 External Clock Drive

2.7V 4.5V 5.5V

Safe Operating Area

16 MHz

8 MHz

VIL1

VIH1

Figure 28-3. External Clock Drive

Symbol Parameter

VCC = 2.7V to 5.5V VCC = 4.5V to 5.5V

UnitsMin Max Min Max

1/tCLCL Oscillator Frequency 0 8 0 16 MHz

tCLCL Clock Period 125 62.5 ns

tCHCX High Time 50 25 ns

tCLCX Low Time 50 25 ns

tCLCH Rise Time 1.6 0.5 s

tCHCL Fall Time 1.6 0.5 s

tCLCL

Change in period from one clock cycle to the next

2 2 %


Notes: 1. R should be in the range 3 k - 100 k, and C should be at least 20 pF. The C values given in the table includes pin capacitance. This will vary with package type.

2. The frequency will vary with package type and board layout.

28.5 System and Reset Characteristics

Notes: 1. The Power-on Reset will not work unless the supply voltage has been below VPOT (falling).

2. VBOT may be below nominal minimum operating voltage for some devices. For devices where this is the case, the device is tested down to VCC = VBOT during the production test. This guarantees that a Brown-out Reset will occur before VCC drops to a voltage where correct operation of the microcontroller is no longer guaranteed. The test is performed using BODLEVEL = 1 and BODLEVEL = 0 for ATmega32A.

28.6 Two-wire Serial Interface CharacteristicsTable 27-2 describes the requirements for devices connected to the Two-wire Serial Bus. The ATmega32A Two-wire SerialInterface meets or exceeds these requirements under the noted conditions.

Figure 28-4. External RC Oscillator, Typical Frequencies (VCC = 5V)

R [k](1) C [pF] f(2)

33 22 650kHz

10 22 2.0MHz

Table 28-1. Reset, Brown-out and Internal Voltage Reference Characteristics


VPOT

Power-on Reset Threshold Voltage (rising)

1.4 2.3 V

Power-on Reset Threshold Voltage (falling)(1) 1.3 2.3 V

VRST RESET Pin Threshold Voltage

0.2 VCC 0.9 VCC V

tRSTMinimum pulse width on RESET Pin

1.5 µs

VBOT

Brown-out Reset Threshold Voltage(2)

BODLEVEL = 1 2.5 2.7 2.9V

BODLEVEL = 0 3.6 4.0 4.2

tBOD

Minimum low voltage period for Brown-out Detection

BODLEVEL = 1 2 µs

BODLEVEL = 0 2 µs

VHYSTBrown-out Detector hysteresis

50 mV

VBG Bandgap reference voltage 1.15 1.23 1.35 V

tBGBandgap reference start-up time

40 70 µs

IBGBandgap reference current consumption

10 µA


Timing symbols refer to Figure 27-5.

Notes: 1. In ATmega32A, this parameter is characterized and not 100% tested.

2. Required only for fSCL > 100kHz.

3. Cb = capacitance of one bus line in pF.

4. fCK = CPU clock frequency

5. This requirement applies to all ATmega32A Two-wire Serial Interface operation. Other devices connected to the Two-wire Serial Bus need only obey the general fSCL requirement.

Table 28-2. Two-wire Serial Bus Requirements

Symbol Parameter Condition Min Max Units

VIL Input Low-voltage -0.5 0.3 VCC V

VIH Input High-voltage 0.7 VCC VCC + 0.5 V

Vhys(1) Hysteresis of Schmitt Trigger Inputs 0.05 VCC

(2) – V

VOL(1) Output Low-voltage 3 mA sink current 0 0.4 V

tr(1) Rise Time for both SDA and SCL 20 + 0.1Cb

(3)(2) 300 ns

tof(1) Output Fall Time from VIHmin to VILmax 10 pF < Cb < 400 pF(3) 20 + 0.1Cb

(3)(2) 250 ns

tSP(1) Spikes Suppressed by Input Filter 0 50(2) ns

Ii Input Current each I/O Pin 0.1 VCC < Vi < 0.9 VCC -10 10 µA

Ci(1) Capacitance for each I/O Pin – 10 pF

fSCL SCL Clock Frequency fCK(4) > max(16fSCL, 250kHz)(5) 0 400 kHz

Rp Value of Pull-up resistor

fSCL 100kHz

fSCL > 100kHz

tHD;STA Hold Time (repeated) START Condition fSCL 100kHz 4.0 – µs

fSCL > 100kHz 0.6 – µs

tLOW Low Period of the SCL Clock fSCL 100kHz 4.7 – µs


tHIGH High period of the SCL clock fSCL 100kHz 4.0 – µs


tSU;STA

Set-up time for a repeated START condition

fSCL 100kHz 4.7 – µs


tHD;DAT Data hold time fSCL 100kHz 0 3.45 µs

fSCL > 100kHz 0 0.9 µs

tSU;DAT Data setup time fSCL 100kHz 250 – ns

fSCL > 100kHz 100 – ns

tSU;STO Setup time for STOP condition fSCL 100kHz 4.0 – µs


tBUF Bus free time between a STOP and START condition

fSCL 100kHz 4.7 – µs


VCC 0.4V–

3 mA---------------------------- 1000ns

Cb-------------------

VCC 0.4V–

3 mA---------------------------- 300ns

Cb----------------


Figure 28-5. Two-wire Serial Bus Timing

28.7 SPI Timing CharacteristicsSee Figure 27-6 and Figure 27-7 for details.

tSU;STA

tLOW

tHIGH

tLOW

tof

tHD;STA tHD;DAT tSU;DATtSU;STO

tBUF

SCL

SDA

tr

Table 28-3. SPI Timing Parameters

Description Mode Min Typ Max

1 SCK period Master See Table 18-4

ns

2 SCK high/low Master 50% duty cycle

3 Rise/Fall time Master 3.6

4 Setup Master 10

5 Hold Master 10

6 Out to SCK Master 0.5 • tSCK

7 SCK to out Master 10

8 SCK to out high Master 10

9 SS low to out Slave 15

10 SCK period Slave 4 • tck

11 SCK high/low Slave 2 • tck

12 Rise/Fall time Slave 1.6 µs

13 Setup Slave 10

ns

14 Hold Slave tck

15 SCK to out Slave 15

16 SCK to SS high Slave 20

17 SS high to tri-state Slave 10

18 SS low to SCK Salve 2 • tck


Figure 28-6. SPI Interface Timing Requirements (Master Mode)

Figure 28-7. SPI Interface Timing Requirements (Slave Mode)

MOSI(Data Output)

SCK(CPOL = 1)

MISO(Data Input)

SCK(CPOL = 0)

SS

MSB LSB

LSBMSB

...

...

6 1

2 2

34 5

87

MISO(Data Output)

SCK(CPOL = 1)

MOSI(Data Input)

SCK(CPOL = 0)

SS

MSB LSB

LSBMSB

...

...

10

11 11

1213 14

1715

9

X

16

18


28.8 ADC Characteristics

Notes: 1. Minimum for AVCC is 2.7V.

2. Maximum for AVCC is 5.5V.

Table 28-4. ADC Characteristics, Single Ended channels, TA = -40C to 85C


Resolution Single Ended Conversion 10 Bits

Absolute Accuracy (Including INL, DNL, Quantization Error, Gain, and Offset Error)

Single Ended ConversionVREF = 4V, VCC = 4VADC clock = 200kHz

1.5 LSB

Single Ended ConversionVREF = 4V, VCC = 4VADC clock = 1MHz

3 LSB

Single Ended ConversionVREF = 4V, VCC = 4VADC clock = 200kHzNoise Reduction mode

1.5 LSB

Single Ended ConversionVREF = 4V, VCC = 4VADC clock = 1MHzNoise Reduction mode

3 LSB

Integral Non-Linearity (INL)Single Ended ConversionVREF = 4V, VCC = 4VADC clock = 200kHz

0.75 LSB

Differential Non-linearity (DNL)Single Ended ConversionVREF = 4V, VCC = 4VADC clock = 200kHz

0.25 LSB

Gain Error Single Ended ConversionVREF = 4V, VCC = 4VADC clock = 200kHz

0.75 LSB

Offset ErrorSingle Ended ConversionVREF = 4V, VCC = 4VADC clock = 200kHz

0.75 LSB

Clock Frequency 50 1000 kHz

Conversion Time 13 260 µs

AVCC Analog Supply Voltage VCC - 0.3(1) VCC + 0.3(2) V

VREF Reference Voltage 2.0 AVCC V

VIN Input voltage GND VREF V

ADC conversion output 0 1023 LSB

Input bandwith 38.5 kHz

VINT Internal Voltage Reference 2.3 2.56 2.7 V

RREF Reference Input Resistance 32 k

RAIN Analog Input Resistance 100 M


Table 28-5. ADC Characteristics, Differential channels, TA = -40C to 85C


Resolution

Gain = 1x 10 Bits

Gain = 10x 10 Bits

Gain = 200x 10 Bits

Absolute Accuracy

Gain = 1xVREF = 4V, VCC = 5VADC clock = 50 - 200kHz

17 LSB


16 LSB


7 LSB

Integral Non-Linearity (INL) (Accuracy after calibration for Offset and Gain Error)


0.75 LSB


0.75 LSB


2 LSB

Gain Error

Gain = 1x 1.6 %

Gain = 10x 1.5 %

Gain = 200x 0.2 %

Offset Error


1 LSB


1.5 LSB


4.5 LSB

Clock Frequency 50 200 kHz

Conversion Time 65 260 µs

AVCC Analog Supply Voltage VCC - 0.3(1) VCC + 0.3(2) V

VREF Reference Voltage 2.0 AVCC - 0.5 V

VIN Input voltage GND AVCC V

VDIFF Input differential voltage -VREF/Gain VREF/Gain/ V

ADC conversion output -511 511 LSB

Input bandwith 4 kHz


Notes: 1. Minimum for AVCC is 2.7V.

2. Maximum for AVCC is 5.5V.

VINT Internal Voltage Reference 2.3 2.56 2.7 V

RREF Reference Input Resistance 32 k

RAIN Analog Input Resistance 100 M

Table 28-5. ADC Characteristics, Differential channels, TA = -40C to 85C (Continued)



29. Typical CharacteristicsThe following charts show typical behavior. These figures are not tested during manufacturing. All current con-sumption measurements are performed with all I/O pins configured as inputs and with internal pull-ups enabled. Asquare wave generator with rail-to-rail output is used as clock source.

The power consumption in Power-down mode is independent of clock selection.

The current consumption is a function of several factors such as: operating voltage, operating frequency, loading ofI/O pins, switching rate of I/O pins, code executed and ambient temperature. The dominating factors are operatingvoltage and frequency.

The current drawn from capacitive loaded pins may be estimated (for one pin) as CL*VCC*f where CL = load capac-itance, VCC = operating voltage and f = average switching frequency of I/O pin.

The parts are characterized at frequencies higher than test limits. Parts are not guaranteed to function properly atfrequencies higher than the ordering code indicates.

The difference between current consumption in Power-down mode with Watchdog Timer enabled and Power-downmode with Watchdog Timer disabled represents the differential current drawn by the Watchdog Timer.

29.1 Active Supply Current

Figure 29-1. Active Supply Current vs. Low Frequency (0.1 - 1.0 MHz)

5.5 V

5.0 V

4.5 V

4.0 V3.6 V3.3 V

2.7 V

0

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1

Frequency (MHz)

I CC (

mA

)


Figure 29-2. Active Supply Current vs. Frequency (1 - 16MHz)

Figure 29-3. Active Supply Current vs. VCC (Internal RC Oscillator, 8MHz)

0

2

4

6

8

10

12

14

16

18

0 2 4 6 8 10 12 14 16

Frequency (MHz)

I CC (

mA

)

2.7V

3.3V

3.6V

4.0V

4.5V

5.0V

5.5V

,

85 °C

25 °C

-40 °C

0

2

4

6

8

10

12

2.5 3 3.5 4 4.5 5 5.5

VCC (V)

I CC (

mA

)




85 °C

25 °C

-40 °C

0

1

2

3

4

5

6

2.5 3 3.5 4 4.5 5 5.5

VCC (V)

I CC (

mA

)

85 °C

25 °C

-40 °C

0

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

2.5 3 3.5 4 4.5 5 5.5

VCC (V)

I CC (

mA

)


Figure 29-6. Active Supply Current vs. VCC (External Oscillator, 32kHz)

29.2 Idle Supply Current

Figure 29-7. Idle Supply Current vs. Low Frequency (0.1 - 1.0MHz)

25 °C

0

20

40

60

80

100

120

140

160

2.5 3 3.5 4 4.5 5 5.5

VCC (V)

I CC (

uA)

5.5 V

5.0 V

4.5 V

4.0 V3.6 V3.3 V

2.7 V

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1

Frequency (MHz)

I CC (

mA

)


Figure 29-8. Idle Supply Current vs. Frequency (1 MHz - 16 MHz)

Figure 29-9. Idle Supply Current vs. VCC (Internal RC Oscillator, 8 MHz)

0

1

2

3

4

5

6

7

8

0 2 4 6 8 10 12 14 16

Frequency (MHz)

I CC (

mA

)

2.7V3.3V

3.6V

4.0V

4.5V

5.0V

5.5V

85 °C

25 °C

-40 °C

0

1

2

3

4

5

2.5 3 3.5 4 4.5 5 5.5

VCC (V)

I CC (

mA

)


Figure 29-10. Idle Supply Current vs. VCC (Internal RC Oscillator, 4MHz)

Figure 29-11. Idle Supply Current vs. VCC (Internal RC Oscillator, 1MHz)

85 °C25 °C

-40 °C

0

0.5

1

1.5

2

2.5

2.5 3 3.5 4 4.5 5 5.5

VCC (V)

I CC (

mA

)

85 °C

25 °C-40 °C

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

2.5 3 3.5 4 4.5 5 5.5

VCC (V)

I CC (

mA

)


Figure 29-12. Idle Supply Current vs. VCC (External Oscillator, 32kHz)

29.3 Power-down Supply Current

Figure 29-13. Power-down Supply Current vs. VCC (Watchdog Timer Disabled)

25 °C

0

5

10

15

20

25

30

35

40

2.5 3 3.5 4 4.5 5 5.5

VCC (V)

I CC (

uA)

85 °C

25 °C

-40 °C

0

0.4

0.8

1.2

1.6

2

2.5 3 3.5 4 4.5 5 5.5

VCC (V)

I CC (

uA)


Figure 29-14. Power-down Supply Current vs. VCC (Watchdog Timer Enabled)

29.4 Power-save Supply Current

Figure 29-15. Power-save Supply Current vs. VCC (Watchdog Timer Disabled)

85 °C

25 °C-40 °C

0

4

8

12

16

20

2.5 3 3.5 4 4.5 5 5.5

VCC (V)

I CC (

uA)

25 °C

0

4

8

12

16

20

2.5 3 3.5 4 4.5 5 5.5

VCC (V)

I CC (

uA)


29.5 Standby Supply Current

Figure 29-16. Standby Supply Current vs. VCC (WDT Disabled)

29.6 Pin Pull-up

Figure 29-17. I/O Pin Pull-up Resistor Current vs. Input Voltage (VCC = 5V)

6MHz_xtal

6MHz_res

4MHz_xtal 4MHz_res

450kHz_res 2MHz_xtal 2MHz_res

1MHz_res

0

0.02

0.04

0.06

0.08

0.1

0.12

0.14

0.16

2.5 3 3.5 4 4.5 5 5.5

VCC (V)

I CC (

mA

)

0

20

40

60

80

100

120

140

0 1 2 3 4 5 6

VOP (V)

I OP (

uA)

85 °C

25 °C

-40 °C


Figure 29-18. I/O Pin Pull-up Resistor Current vs. Input Voltage (VCC = 2.7V)

Figure 29-19. Reset Pull-up Resistor Current vs. Reset Pin Voltage (VCC = 5V)

0

10

20

30

40

50

60

70

0 0.5 1 1.5 2 2.5 3

VOP (V)

I OP

(uA

)

85 °C

25 °C-40 °C

85 °C25 °C

-40 °C

0

20

40

60

80

100

0 1 2 3 4 5 6

VRESET (V)

I RE

SE

T (u

A)


Figure 29-20. Reset Pull-up Resistor Current vs. Reset Pin Voltage (VCC = 2.7V)

29.7 Pin Driver Strength

Figure 29-21. I/O Pin Source Current vs. Output Voltage (VCC = 5V)

85 °C25 °C

-40 °C

0

10

20

30

40

50

60

0 0.5 1 1.5 2 2.5 3

VRESET (V)

I RE

SE

T (u

A)

85 °C

25 °C-40 °C

0

10

20

30

40

50

60

70

80

3 3.4 3.8 4.2 4.6 5

VOH (V)

I OH (

mA

)


Figure 29-22. I/O Pin Source Current vs. Output Voltage (VCC = 3V)

Figure 29-23. I/O Pin Sink Current vs. Output Voltage (VCC = 5V)

85 °C

25 °C

-40 °C

0

5

10

15

20

25

30

35

1 1.5 2 2.5 3

VOH (V)

I OH (

mA

)

85 °C

25 °C

-40 °C

0

10

20

30

40

50

60

70

80

90

0 0.5 1 1.5 2

VOL (V)

I OL

(mA

)


Figure 29-24. I/O Pin Sink Current vs. Output Voltage (VCC = 3V)

29.8 Pin Thresholds and Hysteresis

Figure 29-25. I/O Pin Input Threshold Voltage vs. VCC (VIH, I/O Pin Read as “1”)

85 °C

25 °C

-40 °C

0

5

10

15

20

25

30

35

40

45

0 0.5 1 1.5 2

VOL (V)

I OL

(mA

)

85 °C

25 °C

-40 °C

0

0.5

1

1.5

2

2.5

3

2.5 3 3.5 4 4.5 5 5.5

VCC (V)

Thr

esho

ld (

V)


Figure 29-26. I/O Pin Input Threshold Voltage vs. VCC (VIL, I/O Pin Read as “0”)

Figure 29-27. I/O Pin Input Hysteresis vs. VCC

85 °C

25 °C

-40 °C

0

0.5

1

1.5

2

2.5

2.5 3 3.5 4 4.5 5 5.5

VCC (V)

Thr

esho

ld (

V)

85 °C

25 °C

-40 °C

0

0.2

0.4

0.6

2.5 3 3.5 4 4.5 5 5.5

VCC (V)

Inpu

t Hys

tere

sis

(mV

)


Figure 29-28. Reset Input Threshold Voltage vs. VCC (VIH,Reset Pin Read as “1”)

Figure 29-29. Reset Input Threshold Voltage vs. VCC (VIL,Reset Pin Read as “0”)

85 °C

25 °C

-40 °C

0

0.5

1

1.5

2

2.5

2.5 3 3.5 4 4.5 5 5.5

VCC (V)

Thr

esho

ld (

V)

85 °C25 °C

-40 °C

0

0.5

1

1.5

2

2.5

2.5 3 3.5 4 4.5 5 5.5

VCC (V)

Thr

esho

ld (

V)


Figure 29-30. Reset Input Pin Hysteresis vs. VCC

29.9 BOD Thresholds and Analog Comparator Offset

Figure 29-31. BOD Thresholds vs. Temperature (BOD Level is 4.0V)

85 °C

25 °C

-40 °C

0

0.1

0.2

0.3

0.4

0.5

2.5 3 3.5 4 4.5 5 5.5

VCC (V)

Inpu

t Hys

tere

sis

(mV

)

Rising VCC

Falling VCC

3.8

3.9

4

4.1

-60 -40 -20 0 20 40 60 80 100

Temperature (°C)

Thr

esho

ld (

V)


Figure 29-32. BOD Thresholds vs. Temperature (BOD Level is 2.7V)

Figure 29-33. Bandgap Voltage vs. VCC

Rising VCC

Falling VCC

2.6

2.7

2.8

2.9

-60 -40 -20 0 20 40 60 80 100

Temperature (C)

Thr

esho

ld (

V)

85 °C

25 °C

-40 °C

1.232

1.234

1.236

1.238

1.24

1.242

1.244

1.246

1.248

1.25

2.5 3 3.5 4 4.5 5 5.5

VCC (V)

Ban

dgap

Vol

tage

(V

)


29.10 Internal Oscillator Speed

Figure 29-34. Watchdog Oscillator Frequency vs. VCC

Figure 29-35. Calibrated 8MHz RC Oscillator Frequency vs. Temperature

85 °C

25 °C-40 °C

1120

1140

1160

1180

1200

1220

1240

1260

1280

1300

1320

2.5 3 3.5 4 4.5 5 5.5

VCC (V)

FR

C (

kHz)

5.5 V5.0 V4.5 V

4.0 V

3.6 V3.3 V

2.7 V

6.5

6.7

6.9

7.1

7.3

7.5

7.7

7.9

8.1

8.3

8.5

-60 -40 -20 0 20 40 60 80 100

Temperature

FR

C (

MH

z)


Figure 29-36. Calibrated 8MHz RC Oscillator Frequency vs. VCC

Figure 29-37. Calibrated 8MHz RC Oscillator Frequency vs. Osccal Value

85 °C

25 °C

-40 °C

6

6.5

7

7.5

8

8.5

9

2.5 3 3.5 4 4.5 5 5.5

VCC (V)

FR

C (

MH

z)

85 °C

25 °C

-40 °C

0

2

4

6

8

10

12

14

16

0 16 32 48 64 80 96 112 128 144 160 176 192 208 224 240 256

OSCCAL (X1)

FR

C (

MH

z)




5.5 V

5.0 V4.5 V4.0 V

3.6 V3.3 V

2.7 V

3.5

3.6

3.7

3.8

3.9

4

4.1

4.2

-60 -40 -20 0 20 40 60 80 100

Temperature

FR

C (

MH

z)

85 °C

25 °C

-40 °C

3.5

3.6

3.7

3.8

3.9

4

4.1

4.2

2.5 3 3.5 4 4.5 5 5.5

VCC (V)

FR

C (

MH

z)




85 °C

25 °C

-40 °C

0

1

2

3

4

5

6

7

8

0 16 32 48 64 80 96 112 128 144 160 176 192 208 224 240 256

OSCCAL (X1)

FR

C (

MH

z)

5.5 V5.0 V4.5 V4.0 V3.6 V3.3 V

2.7 V1.8

1.85

1.9

1.95

2

2.05

2.1

-60 -40 -20 0 20 40 60 80 100

Temperature

FR

C (

MH

z)




85 °C

25 °C-40 °C

1.7

1.8

1.9

2

2.1

2.5 3 3.5 4 4.5 5 5.5

VCC (V)

FR

C (

MH

z)

85 °C

25 °C

-40 °C

0

0.5

1

1.5

2

2.5

3

3.5

4

0 16 32 48 64 80 96 112 128 144 160 176 192 208 224 240 256

OSCCAL (X1)

FR

C (

MH

z)




5.5 V5.0 V4.5 V4.0 V3.6 V3.3 V

2.7 V

0.92

0.94

0.96

0.98

1

1.02

1.04

-60 -40 -20 0 20 40 60 80 100

Temperature

FR

C (

MH

z)

85 °C

25 °C-40 °C

0.92

0.94

0.96

0.98

1

1.02

1.04

2.5 3 3.5 4 4.5 5 5.5

VCC (V)

FR

C (

MH

z)



29.11 Current Consumption of Peripheral Units

Figure 29-47. Brownout Detector Current vs. VCC

85 °C25 °C

-40 °C

0

0,2

0,4

0,6

0,8

1

1,2

1,4

1,6

1,8

2

0 16 32 48 64 80 96 112 128 144 160 176 192 208 224 240 256

OSCCAL (X1)

FR

C (

MH

z)

85 °C

25 °C-40 °C

0

2

4

6

8

10

12

14

16

18

20

2.5 3 3.5 4 4.5 5 5.5

VCC (V)

I CC (

uA)


Figure 29-48. ADC Current vs. VCC (AREF = AVCC)

Figure 29-49. AREF External Reference Current vs. VCC

85 °C25 °C

-40 °C

0

50

100

150

200

250

300

350

2.5 3 3.5 4 4.5 5 5.5

VCC (V)

I CC (

uA)

85 °C25 °C

-40 °C

0

50

100

150

200

2.5 3 3.5 4 4.5 5 5.5

VCC (V)

I CC (

uA)


Figure 29-50. Analog Comparator Current vs. VCC

Figure 29-51. Programming Current vs. VCC

85 °C

25 °C-40 °C

20

30

40

50

60

70

80

90

100

2.5 3 3.5 4 4.5 5 5.5

VCC (V)

I CC (

uA)

85 °C

25 °C

-40 °C

0

1

2

3

4

5

6

7

8

9

2.5 3 3.5 4 4.5 5 5.5

VCC (V)

I CC (

mA

)


29.12 Current Consumption in Reset and Reset Pulsewidth

Figure 29-52. Reset Supply Current vs. Low Frequency(0.1 - 1.0MHz, Excluding Current Through The Reset Pull-up)

Figure 29-53. Reset Supply Current vs. Frequency (1 - 16MHz, Excluding Current Through The Reset Pull-up)

5.5 V

5.0 V

4.5 V

4.0 V3.6 V3.3 V

2.7 V

0

0.5

1

1.5

2

2.5

3

0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1

Frequency (MHz)

I CC (

mA

)

0

2

4

6

8

10

12

14

16

0 2 4 6 8 10 12 14 16

Frequency (MHz)

I CC (

mA

)

2.7V

3.3V

3.6V

4.0V

4.5V

5.0V

5.5V


Figure 29-54. Minimum Reset Pulse Width vs. VCC

85 °C

25 °C-40 °C

0

100

200

300

400

500

600

700

800

2.5 3 3.5 4 4.5 5 5.5 6

VCC (V)

Pul

sew

idth

(ns

)


30. Register SummaryAddress Name Bit 7 Bit 6 Bit 5 Bit 4 Bit 3 Bit 2 Bit 1 Bit 0 Page

$3F ($5F) SREG I T H S V N Z C 8

$3E ($5E) SPH – – – – SP11 SP10 SP9 SP8 11

$3D ($5D) SPL SP7 SP6 SP5 SP4 SP3 SP2 SP1 SP0 11

$3C ($5C) OCR0 Timer/Counter0 Output Compare Register 86

$3B ($5B) GICR INT1 INT0 INT2 – – – IVSEL IVCE 48, 71

$3A ($5A) GIFR INTF1 INTF0 INTF2 – – – – – 71

$39 ($59) TIMSK OCIE2 TOIE2 TICIE1 OCIE1A OCIE1B TOIE1 OCIE0 TOIE0 87, 117, 136

$38 ($58) TIFR OCF2 TOV2 ICF1 OCF1A OCF1B TOV1 OCF0 TOV0 87, 117, 136

$37 ($57) SPMCR SPMIE RWWSB – RWWSRE BLBSET PGWRT PGERS SPMEN 264

$36 ($56) TWCR TWINT TWEA TWSTA TWSTO TWWC TWEN – TWIE 202

$35 ($55) MCUCR SE SM2 SM1 SM0 ISC11 ISC10 ISC01 ISC00 36, 69

$34 ($54) MCUCSR JTD ISC2 – JTRF WDRF BORF EXTRF PORF 42, 70, 251

$33 ($53) TCCR0 FOC0 WGM00 COM01 COM00 WGM01 CS02 CS01 CS00 84

$32 ($52) TCNT0 Timer/Counter0 (8 Bits) 86

$31(1) ($51)(1) OSCCAL Oscillator Calibration Register 32

OCDR On-Chip Debug Register 232

$30 ($50) SFIOR ADTS2 ADTS1 ADTS0 – ACME PUD PSR2 PSR10 66,90,137,206,226

$2F ($4F) TCCR1A COM1A1 COM1A0 COM1B1 COM1B0 FOC1A FOC1B WGM11 WGM10 112

$2E ($4E) TCCR1B ICNC1 ICES1 – WGM13 WGM12 CS12 CS11 CS10 114

$2D ($4D) TCNT1H Timer/Counter1 – Counter Register High Byte 116

$2C ($4C) TCNT1L Timer/Counter1 – Counter Register Low Byte 116

$2B ($4B) OCR1AH Timer/Counter1 – Output Compare Register A High Byte 116

$2A ($4A) OCR1AL Timer/Counter1 – Output Compare Register A Low Byte 116

$29 ($49) OCR1BH Timer/Counter1 – Output Compare Register B High Byte 116

$28 ($48) OCR1BL Timer/Counter1 – Output Compare Register B Low Byte 116

$27 ($47) ICR1H Timer/Counter1 – Input Capture Register High Byte 116

$26 ($46) ICR1L Timer/Counter1 – Input Capture Register Low Byte 116

$25 ($45) TCCR2 FOC2 WGM20 COM21 COM20 WGM21 CS22 CS21 CS20 132

$24 ($44) TCNT2 Timer/Counter2 (8 Bits) 135

$23 ($43) OCR2 Timer/Counter2 Output Compare Register 135

$22 ($42) ASSR – – – – AS2 TCN2UB OCR2UB TCR2UB 135

$21 ($41) WDTCR – – – WDTOE WDE WDP2 WDP1 WDP0 43

$20(2) ($40)(2) UBRRH URSEL – – – UBRR[11:8] 171

UCSRC URSEL UMSEL UPM1 UPM0 USBS UCSZ1 UCSZ0 UCPOL 170

$1F ($3F) EEARH – – – – – – EEAR9 EEAR8 20

$1E ($3E) EEARL EEPROM Address Register Low Byte 20

$1D ($3D) EEDR EEPROM Data Register 21

$1C ($3C) EECR – – – – EERIE EEMWE EEWE EERE 21

$1B ($3B) PORTA PORTA7 PORTA6 PORTA5 PORTA4 PORTA3 PORTA2 PORTA1 PORTA0 66

$1A ($3A) DDRA DDA7 DDA6 DDA5 DDA4 DDA3 DDA2 DDA1 DDA0 66

$19 ($39) PINA PINA7 PINA6 PINA5 PINA4 PINA3 PINA2 PINA1 PINA0 66

$18 ($38) PORTB PORTB7 PORTB6 PORTB5 PORTB4 PORTB3 PORTB2 PORTB1 PORTB0 67

$17 ($37) DDRB DDB7 DDB6 DDB5 DDB4 DDB3 DDB2 DDB1 DDB0 67

$16 ($36) PINB PINB7 PINB6 PINB5 PINB4 PINB3 PINB2 PINB1 PINB0 67

$15 ($35) PORTC PORTC7 PORTC6 PORTC5 PORTC4 PORTC3 PORTC2 PORTC1 PORTC0 67

$14 ($34) DDRC DDC7 DDC6 DDC5 DDC4 DDC3 DDC2 DDC1 DDC0 67

$13 ($33) PINC PINC7 PINC6 PINC5 PINC4 PINC3 PINC2 PINC1 PINC0 67

$12 ($32) PORTD PORTD7 PORTD6 PORTD5 PORTD4 PORTD3 PORTD2 PORTD1 PORTD0 67

$11 ($31) DDRD DDD7 DDD6 DDD5 DDD4 DDD3 DDD2 DDD1 DDD0 67

$10 ($30) PIND PIND7 PIND6 PIND5 PIND4 PIND3 PIND2 PIND1 PIND0 68

$0F ($2F) SPDR SPI Data Register 145

$0E ($2E) SPSR SPIF WCOL – – – – – SPI2X 145

$0D ($2D) SPCR SPIE SPE DORD MSTR CPOL CPHA SPR1 SPR0 143

$0C ($2C) UDR USART I/O Data Register 167

$0B ($2B) UCSRA RXC TXC UDRE FE DOR PE U2X MPCM 168

$0A ($2A) UCSRB RXCIE TXCIE UDRIE RXEN TXEN UCSZ2 RXB8 TXB8 169

$09 ($29) UBRRL USART Baud Rate Register Low Byte 171

$08 ($28) ACSR ACD ACBG ACO ACI ACIE ACIC ACIS1 ACIS0 206

$07 ($27) ADMUX REFS1 REFS0 ADLAR MUX4 MUX3 MUX2 MUX1 MUX0 222

$06 ($26) ADCSRA ADEN ADSC ADATE ADIF ADIE ADPS2 ADPS1 ADPS0 224

$05 ($25) ADCH ADC Data Register High Byte 225

$04 ($24) ADCL ADC Data Register Low Byte 225

$03 ($23) TWDR Two-wire Serial Interface Data Register 203

$02 ($22) TWAR TWA6 TWA5 TWA4 TWA3 TWA2 TWA1 TWA0 TWGCE 204

$01 ($21) TWSR TWS7 TWS6 TWS5 TWS4 TWS3 – TWPS1 TWPS0 203

$00 ($20) TWBR Two-wire Serial Interface Bit Rate Register 201


Notes: 1. When the OCDEN Fuse is unprogrammed, the OSCCAL Register is always accessed on this address. Refer to the debug-ger specific documentation for details on how to use the OCDR Register.

2. Refer to the USART description for details on how to access UBRRH and UCSRC.

3. For compatibility with future devices, reserved bits should be written to zero if accessed. Reserved I/O memory addresses should never be written.

4. Some of the Status Flags are cleared by writing a logical one to them. Note that the CBI and SBI instructions will operate on all bits in the I/O Register, writing a one back into any flag read as set, thus clearing the flag. The CBI and SBI instructions work with registers $00 to $1F only.


31. Instruction Set SummaryMnemonics Operands Description Operation Flags #Clocks

ARITHMETIC AND LOGIC INSTRUCTIONS

ADD Rd, Rr Add two Registers Rd Rd + Rr Z,C,N,V,H 1

ADC Rd, Rr Add with Carry two Registers Rd Rd + Rr + C Z,C,N,V,H 1

ADIW Rdl,K Add Immediate to Word Rdh:Rdl Rdh:Rdl + K Z,C,N,V,S 2

SUB Rd, Rr Subtract two Registers Rd Rd - Rr Z,C,N,V,H 1

SUBI Rd, K Subtract Constant from Register Rd Rd - K Z,C,N,V,H 1

SBC Rd, Rr Subtract with Carry two Registers Rd Rd - Rr - C Z,C,N,V,H 1

SBCI Rd, K Subtract with Carry Constant from Reg. Rd Rd - K - C Z,C,N,V,H 1

SBIW Rdl,K Subtract Immediate from Word Rdh:Rdl Rdh:Rdl - K Z,C,N,V,S 2

AND Rd, Rr Logical AND Registers Rd Rd Rr Z,N,V 1

ANDI Rd, K Logical AND Register and Constant Rd Rd K Z,N,V 1

OR Rd, Rr Logical OR Registers Rd Rd v Rr Z,N,V 1

ORI Rd, K Logical OR Register and Constant Rd Rd v K Z,N,V 1

EOR Rd, Rr Exclusive OR Registers Rd Rd Rr Z,N,V 1

COM Rd One’s Complement Rd $FF Rd Z,C,N,V 1

NEG Rd Two’s Complement Rd $00 Rd Z,C,N,V,H 1

SBR Rd,K Set Bit(s) in Register Rd Rd v K Z,N,V 1

CBR Rd,K Clear Bit(s) in Register Rd Rd ($FF - K) Z,N,V 1

INC Rd Increment Rd Rd + 1 Z,N,V 1

DEC Rd Decrement Rd Rd 1 Z,N,V 1

TST Rd Test for Zero or Minus Rd Rd Rd Z,N,V 1

CLR Rd Clear Register Rd Rd Rd Z,N,V 1

SER Rd Set Register Rd $FF None 1

MUL Rd, Rr Multiply Unsigned R1:R0 Rd x Rr Z,C 2

MULS Rd, Rr Multiply Signed R1:R0 Rd x Rr Z,C 2

MULSU Rd, Rr Multiply Signed with Unsigned R1:R0 Rd x Rr Z,C 2

FMUL Rd, Rr Fractional Multiply Unsigned R1:R0 (Rd x Rr) << 1 Z,C 2

FMULS Rd, Rr Fractional Multiply Signed R1:R0 (Rd x Rr) << 1 Z,C 2

FMULSU Rd, Rr Fractional Multiply Signed with Unsigned R1:R0 (Rd x Rr) << 1 Z,C 2

BRANCH INSTRUCTIONS

RJMP k Relative Jump PC PC + k + 1 None 2

IJMP Indirect Jump to (Z) PC Z None 2

JMP k Direct Jump PC k None 3

RCALL k Relative Subroutine Call PC PC + k + 1 None 3

ICALL Indirect Call to (Z) PC Z None 3

CALL k Direct Subroutine Call PC k None 4

RET Subroutine Return PC Stack None 4

RETI Interrupt Return PC Stack I 4

CPSE Rd,Rr Compare, Skip if Equal if (Rd = Rr) PC PC + 2 or 3 None 1 / 2 / 3

CP Rd,Rr Compare Rd Rr Z, N,V,C,H 1

CPC Rd,Rr Compare with Carry Rd Rr C Z, N,V,C,H 1

CPI Rd,K Compare Register with Immediate Rd K Z, N,V,C,H 1

SBRC Rr, b Skip if Bit in Register Cleared if (Rr(b)=0) PC PC + 2 or 3 None 1 / 2 / 3

SBRS Rr, b Skip if Bit in Register is Set if (Rr(b)=1) PC PC + 2 or 3 None 1 / 2 / 3

SBIC P, b Skip if Bit in I/O Register Cleared if (P(b)=0) PC PC + 2 or 3 None 1 / 2 / 3

SBIS P, b Skip if Bit in I/O Register is Set if (P(b)=1) PC PC + 2 or 3 None 1 / 2 / 3

BRBS s, k Branch if Status Flag Set if (SREG(s) = 1) then PCPC+k + 1 None 1 / 2

BRBC s, k Branch if Status Flag Cleared if (SREG(s) = 0) then PCPC+k + 1 None 1 / 2

BREQ k Branch if Equal if (Z = 1) then PC PC + k + 1 None 1 / 2

BRNE k Branch if Not Equal if (Z = 0) then PC PC + k + 1 None 1 / 2

BRCS k Branch if Carry Set if (C = 1) then PC PC + k + 1 None 1 / 2

BRCC k Branch if Carry Cleared if (C = 0) then PC PC + k + 1 None 1 / 2

BRSH k Branch if Same or Higher if (C = 0) then PC PC + k + 1 None 1 / 2

BRLO k Branch if Lower if (C = 1) then PC PC + k + 1 None 1 / 2

BRMI k Branch if Minus if (N = 1) then PC PC + k + 1 None 1 / 2

BRPL k Branch if Plus if (N = 0) then PC PC + k + 1 None 1 / 2

BRGE k Branch if Greater or Equal, Signed if (N V= 0) then PC PC + k + 1 None 1 / 2

BRLT k Branch if Less Than Zero, Signed if (N V= 1) then PC PC + k + 1 None 1 / 2

BRHS k Branch if Half Carry Flag Set if (H = 1) then PC PC + k + 1 None 1 / 2

BRHC k Branch if Half Carry Flag Cleared if (H = 0) then PC PC + k + 1 None 1 / 2

BRTS k Branch if T Flag Set if (T = 1) then PC PC + k + 1 None 1 / 2

BRTC k Branch if T Flag Cleared if (T = 0) then PC PC + k + 1 None 1 / 2

BRVS k Branch if Overflow Flag is Set if (V = 1) then PC PC + k + 1 None 1 / 2


BRVC k Branch if Overflow Flag is Cleared if (V = 0) then PC PC + k + 1 None 1 / 2

BRIE k Branch if Interrupt Enabled if ( I = 1) then PC PC + k + 1 None 1 / 2

BRID k Branch if Interrupt Disabled if ( I = 0) then PC PC + k + 1 None 1 / 2

DATA TRANSFER INSTRUCTIONS

MOV Rd, Rr Move Between Registers Rd Rr None 1

MOVW Rd, Rr Copy Register Word Rd+1:Rd Rr+1:Rr None 1

LDI Rd, K Load Immediate Rd K None 1

LD Rd, X Load Indirect Rd (X) None 2

LD Rd, X+ Load Indirect and Post-Inc. Rd (X), X X + 1 None 2

LD Rd, - X Load Indirect and Pre-Dec. X X - 1, Rd (X) None 2

LD Rd, Y Load Indirect Rd (Y) None 2

LD Rd, Y+ Load Indirect and Post-Inc. Rd (Y), Y Y + 1 None 2

LD Rd, - Y Load Indirect and Pre-Dec. Y Y - 1, Rd (Y) None 2

LDD Rd,Y+q Load Indirect with Displacement Rd (Y + q) None 2

LD Rd, Z Load Indirect Rd (Z) None 2

LD Rd, Z+ Load Indirect and Post-Inc. Rd (Z), Z Z+1 None 2

LD Rd, -Z Load Indirect and Pre-Dec. Z Z - 1, Rd (Z) None 2

LDD Rd, Z+q Load Indirect with Displacement Rd (Z + q) None 2

LDS Rd, k Load Direct from SRAM Rd (k) None 2

ST X, Rr Store Indirect (X) Rr None 2

ST X+, Rr Store Indirect and Post-Inc. (X) Rr, X X + 1 None 2

ST - X, Rr Store Indirect and Pre-Dec. X X - 1, (X) Rr None 2

ST Y, Rr Store Indirect (Y) Rr None 2

ST Y+, Rr Store Indirect and Post-Inc. (Y) Rr, Y Y + 1 None 2

ST - Y, Rr Store Indirect and Pre-Dec. Y Y - 1, (Y) Rr None 2

STD Y+q,Rr Store Indirect with Displacement (Y + q) Rr None 2

ST Z, Rr Store Indirect (Z) Rr None 2

ST Z+, Rr Store Indirect and Post-Inc. (Z) Rr, Z Z + 1 None 2

ST -Z, Rr Store Indirect and Pre-Dec. Z Z - 1, (Z) Rr None 2

STD Z+q,Rr Store Indirect with Displacement (Z + q) Rr None 2

STS k, Rr Store Direct to SRAM (k) Rr None 2

LPM Load Program Memory R0 (Z) None 3

LPM Rd, Z Load Program Memory Rd (Z) None 3

LPM Rd, Z+ Load Program Memory and Post-Inc Rd (Z), Z Z+1 None 3

SPM Store Program Memory (Z) R1:R0 None -

IN Rd, P In Port Rd P None 1

OUT P, Rr Out Port P Rr None 1

PUSH Rr Push Register on Stack Stack Rr None 2

POP Rd Pop Register from Stack Rd Stack None 2

BIT AND BIT-TEST INSTRUCTIONS

SBI P,b Set Bit in I/O Register I/O(P,b) 1 None 2

CBI P,b Clear Bit in I/O Register I/O(P,b) 0 None 2

LSL Rd Logical Shift Left Rd(n+1) Rd(n), Rd(0) 0 Z,C,N,V 1

LSR Rd Logical Shift Right Rd(n) Rd(n+1), Rd(7) 0 Z,C,N,V 1

ROL Rd Rotate Left Through Carry Rd(0)C,Rd(n+1) Rd(n),CRd(7) Z,C,N,V 1

ROR Rd Rotate Right Through Carry Rd(7)C,Rd(n) Rd(n+1),CRd(0) Z,C,N,V 1

ASR Rd Arithmetic Shift Right Rd(n) Rd(n+1), n=0:6 Z,C,N,V 1

SWAP Rd Swap Nibbles Rd(3:0)Rd(7:4),Rd(7:4)Rd(3:0) None 1

BSET s Flag Set SREG(s) 1 SREG(s) 1

BCLR s Flag Clear SREG(s) 0 SREG(s) 1

BST Rr, b Bit Store from Register to T T Rr(b) T 1

BLD Rd, b Bit load from T to Register Rd(b) T None 1

SEC Set Carry C 1 C 1

CLC Clear Carry C 0 C 1

SEN Set Negative Flag N 1 N 1

CLN Clear Negative Flag N 0 N 1

SEZ Set Zero Flag Z 1 Z 1

CLZ Clear Zero Flag Z 0 Z 1

SEI Global Interrupt Enable I 1 I 1

CLI Global Interrupt Disable I 0 I 1

SES Set Signed Test Flag S 1 S 1

CLS Clear Signed Test Flag S 0 S 1

SEV Set Twos Complement Overflow. V 1 V 1

CLV Clear Twos Complement Overflow V 0 V 1SET Set T in SREG T 1 T 1CLT Clear T in SREG T 0 T 1SEH Set Half Carry Flag in SREG H 1 H 1CLH Clear Half Carry Flag in SREG H 0 H 1

Mnemonics Operands Description Operation Flags #Clocks


MCU CONTROL INSTRUCTIONSNOP No Operation None 1SLEEP Sleep (see specific descr. for Sleep function) None 1WDR Watchdog Reset (see specific descr. for WDR/timer) None 1BREAK Break For On-Chip Debug Only None N/A

Mnemonics Operands Description Operation Flags #Clocks


32. Ordering Information

Notes: 1. This device can also be supplied in wafer form. Please contact your local Atmel sales office for detailed ordering information and minimum quantities.

2. Pb-free packaging complies to the European Directive for Restriction of Hazardous Substances (RoHS directive). Also Halide free and fully Green.

3. Tape & Reel

4. See Appendix A ATmega32A 105°C

Speed (MHz) Power Supply Ordering Code(2) Package(1) Operational Range

16 2.7V - 5.5V

ATmega32A-AUSATmega32A-AUR(3)

ATmega32A-PUATmega32A-MUATmega32A-MUR(3)

44A44A40P644M144M1

Industrial(-40oC to 85oC)

ATmega32A-ANATmega32A-ANR(3)

ATmega32A-MNATmega32A-MNR(3)

44A44A44M144M1

Extended(-40oC to 105oC)(4)

Package Type

44A 44-lead, 10 × 10 × 1.0mm, Thin Profile Plastic Quad Flat Package (TQFP)

40P6 40-pin, 0.600” Wide, Plastic Dual Inline Package (PDIP)

44M1 44-pad, 7 × 7 × 1.0mm, Quad Flat No-Lead/Micro Lead Frame Package (QFN/MLF)


33. Packaging Information

33.1 44A

44A, 44-lead, 10 x 10mm body size, 1.0mm body thickness,0.8 mm lead pitch, thin profile plastic quad flat package (TQFP)

C44A

06/02/2014

PIN 1 IDENTIFIER

0°~7°

PIN 1

L

C

A1 A2 A

D1

D

e

E1 E

B

COMMON DIMENSIONS(Unit of Measure = mm)

SYMBOL MIN NOM MAX NOTE

Notes: 1. This package conforms to JEDEC reference MS-026, Variation ACB. 2. Dimensions D1 and E1 do not include mold protrusion. Allowable protrusion is 0.25mm per side. Dimensions D1 and E1 are maximum plastic body size dimensions including mold mismatch. 3. Lead coplanarity is 0.10mm maximum.

A – – 1.20

A1 0.05 – 0.15

A2 0.95 1.00 1.05

D 11.75 12.00 12.25

D1 9.90 10.00 10.10 Note 2

E 11.75 12.00 12.25

E1 9.90 10.00 10.10 Note 2

B 0.30 0.37 0.45

C 0.09 (0.17) 0.20

L 0.45 0.60 0.75

e 0.80 TYP


33.2 40P6

PIN1

E1

A1

B

REF

E

B1

C

L

SEATING PLANE

A

0º ~ 15º

D

e

eB



A – – 4.826

A1 0.381 – –

D 52.070 – 52.578 Note 2

E 15.240 – 15.875

E1 13.462 – 13.970 Note 2

B 0.356 – 0.559

B1 1.041 – 1.651

L 3.048 – 3.556

C 0.203 – 0.381

eB 15.494 – 17.526

e 2.540 TYP

1. This package conforms to JEDEC reference MS-011, Variation AC.2. Dimensions D and E1 do not include mold Flash or Protrusion. Mold Flash or Protrusion shall not exceed 0.25mm (0.010").

Notes:

40P6, 40-lead (0.600"/15.24mm Wide) Plastic Dual Inline Package (PDIP) 40P6 C

13/02/2014


33.3 44M1

TITLE DRAWING NO.GPC REV. Package Drawing Contact: [email protected] 44M1ZWS H

44M1, 44-pad, 7 x 7 x 1.0mm body, lead pitch 0.50mm, 5.20mm exposed pad, thermally enhanced plastic very thin quad flat no lead package (VQFN)

02/13/2014



A 0.80 0.90 1.00

A1 – 0.02 0.05

A3 0.20 REF

b 0.18 0.23 0.30

D

D2 5.00 5.20 5.40

6.90 7.00 7.10

6.90 7.00 7.10

E

E2 5.00 5.20 5.40

e 0.50 BSC

L 0.59 0.64 0.69

K 0.20 0.26 0.41Note: JEDEC Standard MO-220, Fig. 1 (SAW Singulation) VKKD-3.

TOP VIEW

SIDE VIEW

BOTTOM VIEW

D

E

Marked Pin# 1 ID

E2

D2

b e

Pin #1 CornerL

A1

A3

A

SEATING PLANE

Pin #1 Triangle

Pin #1 Chamfer(C 0.30)

Option A

Option B

Pin #1 Notch(0.20 R)

Option C

K

K

123


34. Errata

34.1 ATmega32A, rev. J to rev. K• First Analog Comparator conversion may be delayed• Interrupts may be lost when writing the timer registers in the asynchronous timer• IDCODE masks data from TDI input• Reading EEPROM by using ST or STS to set EERE bit triggers unexpected interrupt request.

1. First Analog Comparator conversion may be delayedIf the device is powered by a slow rising VCC, the first Analog Comparator conversion will take longer thanexpected on some devices.

Problem Fix/WorkaroundWhen the device has been powered or reset, disable then enable the Analog Comparator before the firstconversion.

2. Interrupts may be lost when writing the timer registers in the asynchronous timerThe interrupt will be lost if a timer register that is synchronous timer clock is written when the asynchronousTimer/Counter register (TCNTx) is 0x00.

Problem Fix/WorkaroundAlways check that the asynchronous Timer/Counter register neither have the value 0xFF nor 0x00 before writ-ing to the asynchronous Timer Control Register (TCCRx), asynchronous Timer Counter Register (TCNTx), orasynchronous Output Compare Register (OCRx).

3. IDCODE masks data from TDI inputThe JTAG instruction IDCODE is not working correctly. Data to succeeding devices are replaced by all-onesduring Update-DR.

Problem Fix / Workaround

– If ATmega32A is the only device in the scan chain, the problem is not visible.

– Select the Device ID Register of the ATmega32A by issuing the IDCODE instruction or by entering the Test-Logic-Reset state of the TAP controller to read out the contents of its Device ID Register and possibly data from succeeding devices of the scan chain. Issue the BYPASS instruction to the ATmega32A while reading the Device ID Registers of preceding devices of the boundary scan chain.

– If the Device IDs of all devices in the boundary scan chain must be captured simultaneously, the ATmega32A must be the fist device in the chain.

4. Reading EEPROM by using ST or STS to set EERE bit triggers unexpected interrupt request.Reading EEPROM by using the ST or STS command to set the EERE bit in the EECR register triggers anunexpected EEPROM interrupt request.

Problem Fix / WorkaroundAlways use OUT or SBI to set EERE in EECR.


34.2 ATmega32A, rev. G to rev. I• First Analog Comparator conversion may be delayed• Interrupts may be lost when writing the timer registers in the asynchronous timer• IDCODE masks data from TDI input• Reading EEPROM by using ST or STS to set EERE bit triggers unexpected interrupt request.

1. First Analog Comparator conversion may be delayedIf the device is powered by a slow rising VCC, the first Analog Comparator conversion will take longer thanexpected on some devices.

Problem Fix/WorkaroundWhen the device has been powered or reset, disable then enable the Analog Comparator before the firstconversion.

2. Interrupts may be lost when writing the timer registers in the asynchronous timerThe interrupt will be lost if a timer register that is synchronous timer clock is written when the asynchronousTimer/Counter register (TCNTx) is 0x00.

Problem Fix/WorkaroundAlways check that the asynchronous Timer/Counter register neither have the value 0xFF nor 0x00 before writ-ing to the asynchronous Timer Control Register (TCCRx), asynchronous Timer Counter Register (TCNTx), orasynchronous Output Compare Register (OCRx).

3. IDCODE masks data from TDI inputThe JTAG instruction IDCODE is not working correctly. Data to succeeding devices are replaced by all-onesduring Update-DR.

Problem Fix / Workaround

– If ATmega32A is the only device in the scan chain, the problem is not visible.

– Select the Device ID Register of the ATmega32A by issuing the IDCODE instruction or by entering the Test-Logic-Reset state of the TAP controller to read out the contents of its Device ID Register and possibly data from succeeding devices of the scan chain. Issue the BYPASS instruction to the ATmega32A while reading the Device ID Registers of preceding devices of the boundary scan chain.

– If the Device IDs of all devices in the boundary scan chain must be captured simultaneously, the ATmega32A must be the fist device in the chain.

4. Reading EEPROM by using ST or STS to set EERE bit triggers unexpected interrupt request.Reading EEPROM by using the ST or STS command to set the EERE bit in the EECR register triggers anunexpected EEPROM interrupt request.

Problem Fix / WorkaroundAlways use OUT or SBI to set EERE in EECR.


35. Datasheet Revision HistoryPlease note that the referring page numbers in this section are referred to this document. The referring revision inthis section are referring to the document revision.

35.1 Rev. 8155E – 02/2014

35.2 Rev. 8155D – 10/2013

35.3 Rev. 8155C – 02/2011

35.4 Rev. 8155B – 07/2009

35.5 Rev. 8155A – 06/2008

1. Updated the “Features” with “Capacitive touch sensing” capability.

2. Added “Errata” “ATmega32A, rev. J to rev. K” on page 325.

1.Added nominal values for symbol B, C and L in the TQFP-44 package drawing, “44A” on page 322.

1. Updated the datashee according to the Atmel new brand style guide (new logo, last page, etc).

2. Inserted note in “Performing Page Erase by SPM” on page 241 .

3.Note 6 and Note 7 below Table 28-2, “Two-wire Serial Bus Requirements,” on page 282 have been removed.

4. Updated “Ordering Information” on page 321 to include Tape & Reel and 105°C devices.

5. Updated all “Typical Characteristics” .

1. Updated “Errata” on page 325.

2. Updated the last page with Atmel’s new addresses.

1. Initial revision (Based on the ATmega32/L datasheet 2503N-AVR-06/08)

Changes done compared ATmega32/L datasheet 2503N-AVR-06/08:

- Updated description in “Stack Pointer” on page 11.- All Electrical characteristics is moved to “Electrical Characteristics” on page 296.- Register descriptions are moved to sub sections at the end of each chapter.- Test limits of Reset Pull-up Resistor (RRST) in “DC Characteristics” on page 296.- New graphs in “Typical Characteristics” on page 306.- New “Ordering Information” on page 339.


Table of Contents

Features .....................................................................................................1

1 Pin Configurations ................................................................................... 2

2 Overview ................................................................................................... 3

2.1 Block Diagram ...................................................................................................3

2.2 Pin Descriptions .................................................................................................4

3 Resources ................................................................................................. 5

4 Data Retention .......................................................................................... 5

5 About Code Examples ............................................................................. 6

6 Capacitive touch sensing ........................................................................ 6

7 AVR CPU Core .......................................................................................... 7

7.1 Overview ............................................................................................................7

7.2 ALU – Arithmetic Logic Unit ...............................................................................8

7.3 Status Register ..................................................................................................8

7.4 General Purpose Register File ..........................................................................9

7.5 Stack Pointer ...................................................................................................11

7.6 Instruction Execution Timing ...........................................................................12

7.7 Reset and Interrupt Handling ...........................................................................13

8 AVR Memories ........................................................................................ 15

8.1 Overview ..........................................................................................................15

8.2 In-System Reprogrammable Flash Program Memory .....................................15

8.3 SRAM Data Memory ........................................................................................16

8.4 EEPROM Data Memory ..................................................................................17

8.5 I/O Memory ......................................................................................................18

8.6 Register Description ........................................................................................19

9 System Clock and Clock Options .........................................................23

9.1 Clock Systems and their Distribution ...............................................................23

9.2 Clock Sources .................................................................................................24

9.3 Default Clock Source .......................................................................................24

9.4 Crystal Oscillator .............................................................................................24

9.5 Low-frequency Crystal Oscillator .....................................................................26

9.6 External RC Oscillator .....................................................................................26

9.7 Calibrated Internal RC Oscillator .....................................................................27

iATmega32A [DATASHEET]Atmel-8155D-AVR-ATmega32A-Datasheet_02/2014

9.8 External Clock .................................................................................................28

9.9 Timer/Counter Oscillator ..................................................................................29


10 Power Management and Sleep Modes ................................................. 31

10.1 Sleep Modes ....................................................................................................31

10.2 Idle Mode .........................................................................................................31

10.3 ADC Noise Reduction Mode ............................................................................32

10.4 Power-down Mode ...........................................................................................32

10.5 Power-save Mode ............................................................................................32

10.6 Standby Mode .................................................................................................32

10.7 Extended Standby Mode .................................................................................33

10.8 Minimizing Power Consumption ......................................................................33


11 System Control and Reset ..................................................................... 36

11.1 Resetting the AVR ...........................................................................................36

11.2 Reset Sources .................................................................................................36

11.3 Internal Voltage Reference ..............................................................................39

11.4 Watchdog Timer ..............................................................................................40


12 Interrupts ................................................................................................. 43

12.1 Interrupt Vectors in ATmega32A .....................................................................43


13 I/O Ports .................................................................................................. 48

13.1 Overview ..........................................................................................................48

13.2 Ports as General Digital I/O .............................................................................49

13.3 Alternate Port Functions ..................................................................................53


14 External Interrupts ................................................................................. 65


15 8-bit Timer/Counter0 with PWM ............................................................ 68

15.1 Features ..........................................................................................................68

15.2 Overview ..........................................................................................................68

15.3 Timer/Counter Clock Sources .........................................................................69

15.4 Counter Unit ....................................................................................................69

15.5 Output Compare Unit .......................................................................................70

iiATmega32A [DATASHEET]Atmel-8155D-AVR-ATmega32A-Datasheet_02/2014

15.6 Compare Match Output Unit ............................................................................71

15.7 Modes of Operation .........................................................................................72

15.8 Timer/Counter Timing Diagrams ......................................................................76


16 Timer/Counter0 and Timer/Counter1 Prescalers ................................. 82

16.1 Overview ..........................................................................................................82

16.2 Internal Clock Source ......................................................................................82

16.3 Prescaler Reset ...............................................................................................82

16.4 External Clock Source .....................................................................................82


17 16-bit Timer/Counter1 ............................................................................ 85

17.1 Features ..........................................................................................................85

17.2 Overview ..........................................................................................................85

17.3 Accessing 16-bit Registers ..............................................................................87

17.4 Timer/Counter Clock Sources .........................................................................90

17.5 Counter Unit ....................................................................................................90

17.6 Input Capture Unit ...........................................................................................91

17.7 Compare Match Output Unit ............................................................................95

17.8 Modes of Operation .........................................................................................96

17.9 Timer/Counter Timing Diagrams ....................................................................103

17.10 Register Description ......................................................................................104

18 8-bit Timer/Counter2 with PWM and Asynchronous Operation ....... 111

18.1 Features ........................................................................................................111

18.2 Overview ........................................................................................................111

18.3 Timer/Counter Clock Sources .......................................................................112

18.4 Counter Unit ..................................................................................................112

18.5 Output Compare Unit .....................................................................................113

18.6 Compare Match Output Unit ..........................................................................115

18.7 Modes of Operation .......................................................................................115

18.8 Timer/Counter Timing Diagrams ....................................................................119

18.9 Asynchronous Operation of the Timer/Counter .............................................121

18.10 Timer/Counter Prescaler ...............................................................................123


19 SPI – Serial Peripheral Interface ......................................................... 128

19.1 Features ........................................................................................................128

iiiATmega32A [DATASHEET]Atmel-8155D-AVR-ATmega32A-Datasheet_02/2014

19.2 Overview ........................................................................................................128

19.3 SS Pin Functionality ......................................................................................132

19.4 Data Modes ...................................................................................................134

20 USART ................................................................................................... 136

20.1 Features ........................................................................................................136

20.2 Overview ........................................................................................................136

20.3 Clock Generation ...........................................................................................138

20.4 Frame Formats ..............................................................................................140

20.5 USART Initialization .......................................................................................141

20.6 Data Transmission – The USART Transmitter ..............................................142

20.7 Data Reception – The USART Receiver .......................................................145

20.8 Asynchronous Data Reception ......................................................................149

20.9 Multi-processor Communication Mode ..........................................................152

20.10 Accessing UBRRH/ UCSRC Registers ..........................................................153


20.12 Examples of Baud Rate Setting .....................................................................159

21 Two-wire Serial Interface ..................................................................... 164

21.1 Features ........................................................................................................164

21.2 Two-wire Serial Interface Bus Definition ........................................................164

21.3 Data Transfer and Frame Format ..................................................................165

21.4 Multi-master Bus Systems, Arbitration and Synchronization .........................167

21.5 Overview of the TWI Module .........................................................................169

21.6 Using the TWI ................................................................................................171

21.7 Transmission Modes .....................................................................................173

21.8 Multi-master Systems and Arbitration ............................................................184


22 Analog Comparator .............................................................................. 190

22.1 Overview ........................................................................................................190

22.2 Analog Comparator Multiplexed Input ...........................................................190


23 Analog to Digital Converter ................................................................. 193

23.1 Features ........................................................................................................193

23.2 Overview ........................................................................................................193

23.3 Operation .......................................................................................................194

23.4 Starting a Conversion ....................................................................................195

ivATmega32A [DATASHEET]Atmel-8155D-AVR-ATmega32A-Datasheet_02/2014

23.5 Prescaling and Conversion Timing ................................................................196

23.6 Changing Channel or Reference Selection ...................................................199

23.7 ADC Noise Canceler .....................................................................................200

23.8 ADC Conversion Result .................................................................................203


24 JTAG Interface and On-chip Debug System ...................................... 210

24.1 Features ........................................................................................................210

24.2 Overview ........................................................................................................210

24.3 TAP – Test Access Port ................................................................................210

24.4 TAP Controller ...............................................................................................212

24.5 Using the Boundary-scan Chain ....................................................................213

24.6 Using the On-chip Debug System .................................................................213

24.7 On-chip Debug Specific JTAG Instructions ...................................................214

24.8 Using the JTAG Programming Capabilities ...................................................214


24.10 Bibliography ...................................................................................................215

25 IEEE 1149.1 (JTAG) Boundary-scan ................................................... 216

25.1 Features ........................................................................................................216

25.2 Overview ........................................................................................................216

25.3 Data Registers ...............................................................................................216

25.4 Boundary-scan Specific JTAG Instructions ...................................................218

25.5 Boundary-scan Chain ....................................................................................219

25.6 ATmega32A Boundary-scan Order ...............................................................229

25.7 Boundary-scan Description Language Files ..................................................234


26 Boot Loader Support – Read-While-Write Self-Programming ......... 235

26.1 Features ........................................................................................................235

26.2 Overview ........................................................................................................235

26.3 Application and Boot Loader Flash Sections .................................................235

26.4 Read-While-Write and no Read-While-Write Flash Sections ........................235

26.5 Boot Loader Lock Bits ...................................................................................238

26.6 Entering the Boot Loader Program ................................................................239

26.7 Addressing the Flash during Self-Programming ............................................240

26.8 Self-Programming the Flash ..........................................................................241


vATmega32A [DATASHEET]Atmel-8155D-AVR-ATmega32A-Datasheet_02/2014

27 Memory Programming ......................................................................... 248

27.1 Program And Data Memory Lock Bits ...........................................................248

27.2 Fuse Bits ........................................................................................................249

27.3 Signature Bytes .............................................................................................250

27.4 Calibration Byte .............................................................................................250

27.5 Page Size ......................................................................................................251

27.6 Parallel Programming Parameters, Pin Mapping, and Commands ...............251

27.7 Parallel Programming ....................................................................................253

27.8 SPI Serial Downloading .................................................................................261

27.9 SPI Serial Programming Pin Mapping ...........................................................261

27.10 Programming via the JTAG Interface ............................................................266

28 Electrical Characteristics .................................................................... 278

28.1 Absolute Maximum Ratings* .........................................................................278

28.2 DC Characteristics ........................................................................................278

28.3 Speed Grades ...............................................................................................280

28.4 Clock Characteristics .....................................................................................280

28.5 System and Reset Characteristics ................................................................281

28.6 Two-wire Serial Interface Characteristics ......................................................281

28.7 SPI Timing Characteristics ............................................................................283

28.8 ADC Characteristics ......................................................................................285

29 Typical Characteristics ........................................................................ 288

29.1 Active Supply Current ....................................................................................288

29.2 Idle Supply Current ........................................................................................291

29.3 Power-down Supply Current ..........................................................................294

29.4 Power-save Supply Current ...........................................................................295

29.5 Standby Supply Current ................................................................................296

29.6 Pin Pull-up .....................................................................................................296

29.7 Pin Driver Strength ........................................................................................298

29.8 Pin Thresholds and Hysteresis ......................................................................300

29.9 BOD Thresholds and Analog Comparator Offset ..........................................303

29.10 Internal Oscillator Speed ...............................................................................305

29.11 Current Consumption of Peripheral Units ......................................................311

29.12 Current Consumption in Reset and Reset Pulsewidth ...................................314

30 Register Summary ................................................................................ 316

31 Instruction Set Summary ..................................................................... 318

viATmega32A [DATASHEET]Atmel-8155D-AVR-ATmega32A-Datasheet_02/2014

32 Ordering Information ........................................................................... 321

33 Packaging Information ......................................................................... 322

33.1 44A ................................................................................................................322

33.2 40P6 ..............................................................................................................323

33.3 44M1 ..............................................................................................................324

34 Errata ..................................................................................................... 325

34.1 ATmega32A, rev. J to rev. K .........................................................................325

34.2 ATmega32A, rev. G to rev. I ..........................................................................326

35 Datasheet Revision History ................................................................. 327

35.1 Rev. 8155E – 02/2014 ...................................................................................327

35.2 Rev. 8155D – 10/2013 ...................................................................................327

35.3 Rev. 8155C – 02/2011 ...................................................................................327

35.4 Rev. 8155B – 07/2009 ...................................................................................327

35.5 Rev. 8155A – 06/2008 ...................................................................................327

Table of Contents ....................................................................................... i

viiATmega32A [DATASHEET]Atmel-8155D-AVR-ATmega32A-Datasheet_02/2014

XX X XX XAtmel Corporation 1600 Technology Drive, San Jose, CA 95110 USA T: (+1)(408) 441.0311 F: (+1)(408) 436.4200 | www.atmel.com

© 2014 Atmel Corporation. / Rev.: Atmel-8155D-AVR-ATmega32A-Datasheet_02/2014.

Atmel®, Atmel logo and combinations thereof, and others are registered trademarks or trademarks of Atmel Corporation or its subsidiaries. Other terms and product names may be trademarks of others.

DISCLAIMER: The information in this document is provided in connection with Atmel products. No license, express or implied, by estoppel or otherwise, to any intellectual property rightis granted by this document or in connection with the sale of Atmel products. EXCEPT AS SET FORTH IN THE ATMEL TERMS AND CONDITIONS OF SALES LOCATED ON THEATMEL WEBSITE, ATMEL ASSUMES NO LIABILITY WHATSOEVER AND DISCLAIMS ANY EXPRESS, IMPLIED OR STATUTORY WARRANTY RELATING TO ITS PRODUCTSINCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTY OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE, OR NON-INFRINGEMENT. IN NO EVENTSHALL ATMEL BE LIABLE FOR ANY DIRECT, INDIRECT, CONSEQUENTIAL, PUNITIVE, SPECIAL OR INCIDENTAL DAMAGES (INCLUDING, WITHOUT LIMITATION, DAMAGESFOR LOSS AND PROFITS, BUSINESS INTERRUPTION, OR LOSS OF INFORMATION) ARISING OUT OF THE USE OR INABILITY TO USE THIS DOCUMENT, EVEN IF ATMEL HASBEEN ADVISED OF THE POSSIBILITY OF SUCH DAMAGES. Atmel makes no representations or warranties with respect to the accuracy or completeness of the contents of thisdocument and reserves the right to make changes to specifications and products descriptions at any time without notice. Atmel does not make any commitment to update the informationcontained herein. Unless specifically provided otherwise, Atmel products are not suitable for, and shall not be used in, automotive applications. Atmel products are not intended,authorized, or warranted for use as components in applications intended to support or sustain life.

SAFETY-CRITICAL, MILITARY, AND AUTOMOTIVE APPLICATIONS DISCLAIMER: Atmel products are not designed for and will not be used in connection with any applications wherethe failure of such products would reasonably be expected to result in significant personal injury or death (“Safety-Critical Applications”) without an Atmel officer's specific writtenconsent. Safety-Critical Applications include, without limitation, life support devices and systems, equipment or systems for the operation of nuclear facilities and weapons systems.Atmel products are not designed nor intended for use in military or aerospace applications or environments unless specifically designated by Atmel as military-grade. Atmel products arenot designed nor intended for use in automotive applications unless specifically designated by Atmel as automotive-grade.

https://plus.google.com/117391618085377601886/posts

https://twitter.com/Atmel

http://www.linkedin.com/company/atmel-corporation

http://www.youtube.com/user/AtmelCorporation

https://www.facebook.com/AtmelCorporation

http://en.wikipedia.org/wiki/Atmel

www.atmel.com

www.atmel.com

Download ATmega32 - ickala.com

Documents

Transcript of Download ATmega32 - ickala.com